rig-core 0.46.0

{
"anthropic": {"models": {
  "claude-fable-5": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"release_date":"2026-06-07","last_updated":"2026-06-09","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-fable-5-1": {"name":"Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never"}},
  "claude-haiku-4-5": {"name":"Claude Haiku 4.5 (latest)","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-haiku-4-5-20251001": {"name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-haiku-5-5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false},
  "claude-opus-4-0": {"name":"Claude Opus 4","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","pdf"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","status"],"prefill":true}},
  "claude-opus-4-1": {"name":"Claude Opus 4.1","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","pdf"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","status"],"prefill":true}},
  "claude-opus-4-1-20250805": {"name":"Claude Opus 4.1","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","pdf"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","status"],"prefill":true}},
  "claude-opus-4-20250514": {"name":"Claude Opus 4","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","pdf"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","status"],"prefill":true}},
  "claude-opus-4-5": {"name":"Claude Opus 4.5 (latest)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-opus-4-5-20251101": {"name":"Claude Opus 4.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-01","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-opus-4-6": {"name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05-31","release_date":"2026-02-04","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "claude-opus-4-7": {"name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-14","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-opus-4-8": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-opus-5": {"name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-opus-5-5": {"name":"Claude Opus 5.5","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":8,"output":40,"cache_read":0.4,"cache_write":10},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "claude-sonnet-4-0": {"name":"Claude Sonnet 4","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","pdf"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","status"],"prefill":true}},
  "claude-sonnet-4-20250514": {"name":"Claude Sonnet 4","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","pdf"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","status"],"prefill":true}},
  "claude-sonnet-4-5": {"name":"Claude Sonnet 4.5 (latest)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-sonnet-4-5-20250929": {"name":"Claude Sonnet 4.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-sonnet-4-6": {"name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "claude-sonnet-5": {"name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-29","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-sonnet-5-5": {"name":"Claude Sonnet 5.5","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}}
}},
"aws_bedrock": {"models": {
  "amazon.nova-2-lite-v1:0": {"name":"Nova 2 Lite","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova","canonical_model_id":"amazon/nova-2-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":65535},"cost":{"input":0.33,"output":2.75,"cache_read":0.0825,"cache_write":0.33},"knowledge":"2025-10","release_date":"2025-12-02","last_updated":"2025-12-01","open_weights":false},
  "amazon.nova-canvas-v1:0": {"name":"Nova Canvas","modalities":{"input":["text","image"]},"rig":{"pinned":["name","modalities"]}},
  "amazon.nova-lite-v1:0": {"name":"Nova Lite","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-lite","canonical_model_id":"amazon/nova-lite","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.06,"output":0.24,"cache_read":0.015,"cache_write":0.06},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "amazon.nova-micro-v1:0": {"name":"Nova Micro","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-micro","canonical_model_id":"amazon/nova-micro","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":10000},"cost":{"input":0.035,"output":0.14,"cache_read":0.00875,"cache_write":0.035},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "amazon.nova-pro-v1:0": {"name":"Nova Pro","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova-pro","canonical_model_id":"amazon/nova-pro","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.8,"output":3.2,"cache_read":0.2,"cache_write":0.8},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "amazon.titan-embed-image-v1": {"name":"Titan Multimodal Embeddings","modalities":{"input":["text","image"]},"rig":{"pinned":["name","modalities"]}},
  "amazon.titan-embed-text-v1": {"name":"Titan Text Embeddings","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "amazon.titan-embed-text-v2:0": {"name":"Titan Text Embeddings V2","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "anthropic.claude-fable-5": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic.claude-fable-5-1": {"name":"Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never"}},
  "anthropic.claude-haiku-4-5-20251001-v1:0": {"name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic.claude-haiku-5-5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "anthropic.claude-opus-4-1-20250805-v1:0": {"name":"Claude Opus 4.1","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-1-20250805","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic.claude-opus-4-5-20251101-v1:0": {"name":"Claude Opus 4.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5-20251101","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic.claude-opus-4-6-v1": {"name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "anthropic.claude-opus-4-7": {"name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic.claude-opus-4-8": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic.claude-opus-5": {"name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic.claude-opus-5-5": {"name":"Claude Opus 5.5","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "anthropic.claude-sonnet-4-5-20250929-v1:0": {"name":"Claude Sonnet 4.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5-20250929","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic.claude-sonnet-4-6": {"name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "anthropic.claude-sonnet-5": {"name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic.claude-sonnet-5-5": {"name":"Claude Sonnet 5.5","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "apac.amazon.nova-lite-v1:0": {"name":"Nova Lite (APAC)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-lite","canonical_model_id":"amazon/nova-lite","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.063,"output":0.252,"cache_read":0.01575,"cache_write":0.063},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "apac.amazon.nova-micro-v1:0": {"name":"Nova Micro (APAC)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-micro","canonical_model_id":"amazon/nova-micro","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":10000},"cost":{"input":0.037,"output":0.148,"cache_read":0.00925,"cache_write":0.037},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "apac.amazon.nova-pro-v1:0": {"name":"Nova Pro (APAC)","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova-pro","canonical_model_id":"amazon/nova-pro","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.84,"output":3.36,"cache_read":0.21,"cache_write":0.84},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "apac.anthropic.claude-sonnet-4-20250514-v1:0": {"name":"Claude Sonnet 4 (APAC)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-20250514","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "au.anthropic.claude-haiku-4-5-20251001-v1:0": {"name":"Claude Haiku 4.5 (AU)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1.1,"output":5.5,"cache_read":0.11,"cache_write":1.375},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "au.anthropic.claude-haiku-5-5": {"name":"Claude Haiku 5.5 (AU)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.11,"output":0.55,"cache_read":0.011,"cache_write":0.1375,"tiers":[{"input":0.55,"output":2.75,"cache_read":0.055,"cache_write":0.6875,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "au.anthropic.claude-opus-4-6-v1": {"name":"AU Anthropic Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "au.anthropic.claude-opus-4-7": {"name":"Claude Opus 4.7 (AU)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "au.anthropic.claude-opus-4-8": {"name":"Claude Opus 4.8 (AU)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "au.anthropic.claude-opus-5": {"name":"Claude Opus 5 (AU)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "au.anthropic.claude-opus-5-5": {"name":"Claude Opus 5.5 (AU)","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4.4,"output":22,"cache_read":0.22,"cache_write":5.5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "au.anthropic.claude-sonnet-4-5-20250929-v1:0": {"name":"Claude Sonnet 4.5 (AU)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5-20250929","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "au.anthropic.claude-sonnet-4-6": {"name":"AU Anthropic Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "au.anthropic.claude-sonnet-5": {"name":"Claude Sonnet 5 (AU)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.22,"cache_write":2.75},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "au.anthropic.claude-sonnet-5-5": {"name":"Claude Sonnet 5.5 (AU)","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.11,"cache_write":2.75},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "ca.amazon.nova-lite-v1:0": {"name":"Nova Lite (CA)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-lite","canonical_model_id":"amazon/nova-lite","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.064,"output":0.256,"cache_read":0.016,"cache_write":0.064},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "cohere.embed-english-v3": {"name":"Cohere Embed English v3","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "cohere.embed-multilingual-v3": {"name":"Cohere Embed Multilingual v3","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "deepseek.r1-v1:0": {"name":"DeepSeek-R1","description":"Classic open reasoning model for transparent math, coding, and deliberate problem solving","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-r1","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":1.35,"output":5.4},"knowledge":"2024-07","release_date":"2025-01-20","last_updated":"2025-05-29","open_weights":true,"status":"deprecated"},
  "deepseek.v3-v1:0": {"name":"DeepSeek-V3.1","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3.1","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":81920},"cost":{"input":0.58,"output":1.68},"release_date":"2025-08-21","last_updated":"2025-09-18","open_weights":true,"status":"deprecated"},
  "deepseek.v3.2": {"name":"DeepSeek V3.2","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes, sparse attention, and tool-use","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3.2","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":81920},"cost":{"input":0.62,"output":1.85},"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2026-02-06","open_weights":true},
  "eu.amazon.nova-2-lite-v1:0": {"name":"Nova 2 Lite (EU)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova","canonical_model_id":"amazon/nova-2-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":65535},"cost":{"input":0.374,"output":3.157,"cache_read":0.0935,"cache_write":0.374},"knowledge":"2025-10","release_date":"2025-12-02","last_updated":"2025-12-01","open_weights":false},
  "eu.amazon.nova-lite-v1:0": {"name":"Nova Lite (EU)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-lite","canonical_model_id":"amazon/nova-lite","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.069,"output":0.276,"cache_read":0.01725,"cache_write":0.069},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "eu.amazon.nova-micro-v1:0": {"name":"Nova Micro (EU)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-micro","canonical_model_id":"amazon/nova-micro","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":10000},"cost":{"input":0.04,"output":0.16,"cache_read":0.01,"cache_write":0.04},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "eu.amazon.nova-pro-v1:0": {"name":"Nova Pro (EU)","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova-pro","canonical_model_id":"amazon/nova-pro","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.92,"output":3.68,"cache_read":0.23,"cache_write":0.92},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "eu.anthropic.claude-fable-5": {"name":"Claude Fable 5 (EU)","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":11,"output":55,"cache_read":1.1,"cache_write":13.75},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "eu.anthropic.claude-haiku-4-5-20251001-v1:0": {"name":"Claude Haiku 4.5 (EU)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1.1,"output":5.5,"cache_read":0.11,"cache_write":1.375},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "eu.anthropic.claude-haiku-5-5": {"name":"Claude Haiku 5.5 (EU)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.11,"output":0.55,"cache_read":0.011,"cache_write":0.1375,"tiers":[{"input":0.55,"output":2.75,"cache_read":0.055,"cache_write":0.6875,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "eu.anthropic.claude-opus-4-5-20251101-v1:0": {"name":"Claude Opus 4.5 (EU)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5-20251101","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "eu.anthropic.claude-opus-4-6-v1": {"name":"Claude Opus 4.6 (EU)","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "eu.anthropic.claude-opus-4-7": {"name":"Claude Opus 4.7 (EU)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "eu.anthropic.claude-opus-4-8": {"name":"Claude Opus 4.8 (EU)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "eu.anthropic.claude-opus-5": {"name":"Claude Opus 5 (EU)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "eu.anthropic.claude-opus-5-5": {"name":"Claude Opus 5.5 (EU)","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4.4,"output":22,"cache_read":0.22,"cache_write":5.5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "eu.anthropic.claude-sonnet-4-20250514-v1:0": {"name":"Claude Sonnet 4 (EU)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-20250514","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "eu.anthropic.claude-sonnet-4-5-20250929-v1:0": {"name":"Claude Sonnet 4.5 (EU)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5-20250929","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "eu.anthropic.claude-sonnet-4-6": {"name":"Claude Sonnet 4.6 (EU)","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "eu.anthropic.claude-sonnet-5": {"name":"Claude Sonnet 5 (EU)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.22,"cache_write":2.75},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "eu.anthropic.claude-sonnet-5-5": {"name":"Claude Sonnet 5.5 (EU)","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.11,"cache_write":2.75},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "eu.mistral.pixtral-large-2502-v1:0": {"name":"Pixtral Large (25.02) (EU)","description":"Mistral vision-language model for image understanding and multimodal chat","family":"pixtral","canonical_model_id":"mistral/pixtral-large-2502","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":2,"output":6},"release_date":"2025-04-08","last_updated":"2025-04-08","open_weights":false},
  "global.amazon.nova-2-lite-v1:0": {"name":"Nova 2 Lite (Global)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova","canonical_model_id":"amazon/nova-2-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":65535},"cost":{"input":0.3,"output":2.5,"cache_read":0.075,"cache_write":0.3},"knowledge":"2025-10","release_date":"2025-12-02","last_updated":"2025-12-01","open_weights":false},
  "global.anthropic.claude-fable-5": {"name":"Claude Fable 5 (Global)","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "global.anthropic.claude-fable-5-1": {"name":"Claude Fable 5.1 (Global)","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never"}},
  "global.anthropic.claude-haiku-4-5-20251001-v1:0": {"name":"Claude Haiku 4.5 (Global)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "global.anthropic.claude-haiku-5-5": {"name":"Claude Haiku 5.5 (Global)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "global.anthropic.claude-opus-4-5-20251101-v1:0": {"name":"Claude Opus 4.5 (Global)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5-20251101","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "global.anthropic.claude-opus-4-6-v1": {"name":"Claude Opus 4.6 (Global)","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "global.anthropic.claude-opus-4-7": {"name":"Claude Opus 4.7 (Global)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "global.anthropic.claude-opus-4-8": {"name":"Claude Opus 4.8 (Global)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "global.anthropic.claude-opus-5": {"name":"Claude Opus 5 (Global)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "global.anthropic.claude-opus-5-5": {"name":"Claude Opus 5.5 (Global)","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "global.anthropic.claude-sonnet-4-20250514-v1:0": {"name":"Claude Sonnet 4 (Global)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-20250514","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "global.anthropic.claude-sonnet-4-5-20250929-v1:0": {"name":"Claude Sonnet 4.5 (Global)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5-20250929","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "global.anthropic.claude-sonnet-4-6": {"name":"Claude Sonnet 4.6 (Global)","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "global.anthropic.claude-sonnet-5": {"name":"Claude Sonnet 5 (Global)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "global.anthropic.claude-sonnet-5-5": {"name":"Claude Sonnet 5.5 (Global)","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "global.moonshotai.kimi-k3": {"name":"Kimi K3 (Global)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":128000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "global.openai.gpt-5.6-luna": {"name":"GPT-5.6 Luna (Global)","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "global.openai.gpt-5.6-sol": {"name":"GPT-5.6 Sol (Global)","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "global.openai.gpt-5.6-terra": {"name":"GPT-5.6 Terra (Global)","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "global.openai.gpt-6-astra": {"name":"GPT-6 Astra (Global)","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"experimental":{"modes":{"ultrafast":{"cost":{"input":60,"output":300,"cache_read":6,"cache_write":75},"provider":{"body":{"additionalModelRequestFields":{"service_tier":"ultrafast"}}}}}}},
  "global.openai.gpt-6-luna": {"name":"GPT-6 Luna (Global)","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "global.openai.gpt-6-sol": {"name":"GPT-6 Sol (Global)","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "global.openai.gpt-6.1-sol": {"name":"GPT-6.1 Sol (Global)","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"experimental":{"modes":{"ultrafast":{"cost":{"input":12,"output":60,"cache_read":0.6,"cache_write":15},"provider":{"body":{"additionalModelRequestFields":{"service_tier":"ultrafast"}}}}}}},
  "global.openai.gpt-6.1-terra": {"name":"GPT-6.1 Terra (Global)","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false},
  "global.xai.grok-4.6": {"name":"Grok 4.6 (Global)","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-18","open_weights":false},
  "global.xai.grok-4.7": {"name":"Grok 4.7 (Global)","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-28","open_weights":false},
  "global.zai.glm-5.3": {"name":"GLM-5.3 (Global)","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":1.68,"output":5.28,"cache_read":0.312,"cache_write":2.1},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "google.gemma-3-12b-it": {"name":"Gemma 3 12B IT","description":"Open multimodal Gemma instruction model for multilingual text generation and image understanding","family":"gemma","canonical_model_id":"google/gemma-3-12b-it","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":0.09,"output":0.29},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true,"status":"deprecated"},
  "google.gemma-3-27b-it": {"name":"Gemma 3 27B IT","description":"Largest open Gemma 3 instruction model for multilingual text generation and visual understanding","family":"gemma","canonical_model_id":"google/gemma-3-27b-it","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":0.23,"output":0.38},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true,"status":"deprecated"},
  "google.gemma-3-4b-it": {"name":"Gemma 3 4B IT","description":"Open multimodal Gemma instruction model for efficient text generation and image understanding","family":"gemma","canonical_model_id":"google/gemma-3-4b-it","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":4096},"cost":{"input":0.04,"output":0.08},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true,"status":"deprecated"},
  "google.gemma-4-26b-a4b": {"name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-26b-a4b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.13,"output":0.4},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "google.gemma-4-31b": {"name":"Gemma 4 31B IT","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","canonical_model_id":"google/gemma-4-31b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.14,"output":0.4},"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "google.gemma-4-e2b": {"name":"Gemma 4 E2B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-E2B-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":0.04,"output":0.08},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "in.anthropic.claude-haiku-4-5-20251001-v1:0": {"name":"Claude Haiku 4.5 (India)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1.1,"output":5.5,"cache_read":0.11,"cache_write":1.375},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "in.anthropic.claude-opus-5": {"name":"Claude Opus 5 (India)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "in.anthropic.claude-sonnet-5": {"name":"Claude Sonnet 5 (India)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.22,"cache_write":2.75},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "in.openai.gpt-5.6-luna": {"name":"GPT-5.6 Luna (India)","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.22,"output":1.32,"cache_read":0.022,"cache_write":0.275,"tiers":[{"input":0.44,"output":1.98,"cache_read":0.044,"cache_write":0.55,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.44,"output":1.98,"cache_read":0.044,"cache_write":0.55}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "in.openai.gpt-5.6-terra": {"name":"GPT-5.6 Terra (India)","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.2,"output":13.2,"cache_read":0.22,"cache_write":2.75,"tiers":[{"input":4.4,"output":19.8,"cache_read":0.44,"cache_write":5.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4.4,"output":19.8,"cache_read":0.44,"cache_write":5.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "jp.amazon.nova-2-lite-v1:0": {"name":"Nova 2 Lite (JP)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova","canonical_model_id":"amazon/nova-2-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":65535},"cost":{"input":0.396,"output":3.311,"cache_read":0.099,"cache_write":0.396},"knowledge":"2025-10","release_date":"2025-12-02","last_updated":"2025-12-01","open_weights":false},
  "jp.anthropic.claude-haiku-4-5-20251001-v1:0": {"name":"Claude Haiku 4.5 (JP)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1.1,"output":5.5,"cache_read":0.11,"cache_write":1.375},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "jp.anthropic.claude-haiku-5-5": {"name":"Claude Haiku 5.5 (JP)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.11,"output":0.55,"cache_read":0.011,"cache_write":0.1375,"tiers":[{"input":0.55,"output":2.75,"cache_read":0.055,"cache_write":0.6875,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "jp.anthropic.claude-opus-4-7": {"name":"Claude Opus 4.7 (JP)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "jp.anthropic.claude-opus-4-8": {"name":"Claude Opus 4.8 (JP)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "jp.anthropic.claude-opus-5": {"name":"Claude Opus 5 (JP)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "jp.anthropic.claude-opus-5-5": {"name":"Claude Opus 5.5 (JP)","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4.4,"output":22,"cache_read":0.22,"cache_write":5.5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "jp.anthropic.claude-sonnet-4-5-20250929-v1:0": {"name":"Claude Sonnet 4.5 (JP)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5-20250929","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "jp.anthropic.claude-sonnet-4-6": {"name":"Claude Sonnet 4.6 (JP)","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "jp.anthropic.claude-sonnet-5": {"name":"Claude Sonnet 5 (JP)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.22,"cache_write":2.75},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "meta.llama3-1-70b-instruct-v1:0": {"name":"Llama 3.1 70B Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.1-70b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.72,"output":0.72},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true,"status":"deprecated"},
  "meta.llama3-1-8b-instruct-v1:0": {"name":"Llama 3.1 8B Instruct","description":"Compact open Llama model for lightweight chat, drafting, and self-hosting","family":"llama","canonical_model_id":"meta/llama-3.1-8b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.22,"output":0.22},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true,"status":"deprecated"},
  "meta.llama3-3-70b-instruct-v1:0": {"name":"Llama 3.3 70B Instruct","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","canonical_model_id":"meta/llama-3.3-70b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.72,"output":0.72},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true,"status":"deprecated"},
  "meta.llama3-70b-instruct-v1:0": {"name":"Llama 3 70B Instruct","reasoning":false,"tool_call":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":8192},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context"]}},
  "meta.llama3-8b-instruct-v1:0": {"name":"Llama 3 8B Instruct","reasoning":false,"tool_call":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":8192},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context"]}},
  "meta.llama4-maverick-17b-instruct-v1:0": {"name":"Llama 4 Maverick 17B Instruct","description":"Open multimodal Llama for strong reasoning with efficient everyday serving","family":"llama","canonical_model_id":"meta/llama-4-maverick-17b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":8192},"cost":{"input":0.24,"output":0.97},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true,"status":"deprecated"},
  "meta.llama4-scout-17b-instruct-v1:0": {"name":"Llama 4 Scout 17B Instruct","description":"Open Llama with long-context vision for efficient multimodal agents","family":"llama","canonical_model_id":"meta/llama-4-scout-17b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":10000000,"output":8192},"cost":{"input":0.17,"output":0.66},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true,"status":"deprecated"},
  "minimax.minimax-m2": {"name":"MiniMax-M2","description":"Efficient open MiniMax model built for coding agents and tool-heavy workflows","family":"minimax","canonical_model_id":"minimax/MiniMax-M2","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204608,"output":128000},"cost":{"input":0.3,"output":1.2},"release_date":"2025-10-27","last_updated":"2025-10-27","open_weights":true,"status":"deprecated"},
  "minimax.minimax-m2.1": {"name":"MiniMax-M2.1","description":"Earlier MiniMax agent model for practical coding and productivity tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.1","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":196608,"output":131072},"cost":{"input":0.3,"output":1.2},"release_date":"2025-12-23","last_updated":"2025-12-23","open_weights":true,"status":"deprecated"},
  "minimax.minimax-m2.5": {"name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":196608,"output":98304},"cost":{"input":0.3,"output":1.2},"release_date":"2026-02-12","last_updated":"2026-03-18","open_weights":true},
  "mistral.devstral-2-123b": {"name":"Devstral 2 123B","description":"Mistral's coding-agent model for repository work, terminal tasks, and software fixes","family":"devstral","canonical_model_id":"mistral/devstral-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":8192},"cost":{"input":0.4,"output":2},"knowledge":"2025-12","release_date":"2025-12-09","last_updated":"2025-12-09","open_weights":true,"status":"deprecated"},
  "mistral.magistral-small-2509": {"name":"Magistral Small 1.2","description":"Open multimodal reasoning model for transparent analysis of text and images","family":"magistral","canonical_model_id":"mistral/magistral-small-2509","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":40000},"cost":{"input":0.5,"output":1.5},"release_date":"2025-09-18","last_updated":"2025-09-18","open_weights":true},
  "mistral.ministral-3-14b-instruct": {"name":"Ministral 14B 3.0","description":"Open vision-language model for efficient local deployment, instruction following, and tool use","family":"ministral","canonical_model_id":"mistral/ministral-3-14b-instruct-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.2,"output":0.2},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistral.ministral-3-3b-instruct": {"name":"Ministral 3 3B","description":"Compact open vision-language model for edge deployment, instruction following, and tool use","family":"ministral","canonical_model_id":"mistral/ministral-3-3b-instruct-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":8192},"cost":{"input":0.1,"output":0.1},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistral.ministral-3-8b-instruct": {"name":"Ministral 3 8B","description":"Compact open vision-language model for edge deployment, instruction following, and tool use","family":"ministral","canonical_model_id":"mistral/ministral-3-8b-instruct-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.15,"output":0.15},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistral.mistral-7b-instruct-v0:2": {"name":"Mistral 7B Instruct","reasoning":false,"tool_call":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":32768},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context"]}},
  "mistral.mistral-large-2402-v1:0": {"name":"Mistral Large (24.02)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":32768},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context"]}},
  "mistral.mistral-large-3-675b-instruct": {"name":"Mistral Large 3","description":"Mistral's largest general model for enterprise agents, coding, and multilingual reasoning","family":"mistral-large","canonical_model_id":"mistral/mistral-large-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":8192},"cost":{"input":0.5,"output":1.5},"knowledge":"2024-11","release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistral.mistral-small-2402-v1:0": {"name":"Mistral Small (24.02)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":32768},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context"]}},
  "mistral.mixtral-8x7b-instruct-v0:1": {"name":"Mixtral 8x7B Instruct","reasoning":false,"tool_call":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":32768},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context"]}},
  "mistral.pixtral-large-2502-v1:0": {"name":"Pixtral Large (25.02)","description":"Mistral vision-language model for image understanding and multimodal chat","family":"pixtral","canonical_model_id":"mistral/pixtral-large-2502","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":2,"output":6},"release_date":"2025-04-08","last_updated":"2025-04-08","open_weights":false},
  "mistral.voxtral-mini-3b-2507": {"name":"Voxtral Mini 3B 2507","description":"Open audio-language model for speech transcription, audio understanding, and voice-driven tool use","family":"voxtral","canonical_model_id":"mistral/voxtral-mini-3b-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","audio"],"output":["text"]},"limit":{"context":32768,"output":4096},"cost":{"input":0.04,"output":0.04},"release_date":"2025-07-15","last_updated":"2025-07-15","open_weights":true,"status":"deprecated"},
  "mistral.voxtral-small-24b-2507": {"name":"Voxtral Small 24B 2507","description":"Open audio-language model for speech transcription, audio understanding, and voice-driven tool use","family":"voxtral","canonical_model_id":"mistral/voxtral-small-24b-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","audio"],"output":["text"]},"limit":{"context":32768,"output":8192},"cost":{"input":0.1,"output":0.3},"release_date":"2025-07-15","last_updated":"2025-07-15","open_weights":true,"status":"deprecated"},
  "moonshot.kimi-k2-thinking": {"name":"Kimi K2 Thinking","description":"Thinking Kimi model for slower research passes, planning, and hard technical questions","family":"kimi-thinking","canonical_model_id":"moonshotai/kimi-k2-thinking","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":16000},"cost":{"input":0.6,"output":2.5},"knowledge":"2024-08","release_date":"2025-11-06","last_updated":"2025-12-02","open_weights":true,"status":"deprecated"},
  "moonshotai.kimi-k2.5": {"name":"Kimi K2.5","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":0.6,"output":3},"knowledge":"2025-01","release_date":"2026-01-27","last_updated":"2026-02-06","open_weights":true},
  "nvidia.nemotron-nano-12b-v2": {"name":"NVIDIA Nemotron Nano 12B v2 VL BF16","description":"Nemotron multimodal model for visual reasoning and agentic AI workflows","family":"nemotron","canonical_model_id":"nvidia/nemotron-nano-12b-v2-vl","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":0.2,"output":0.6},"release_date":"2025-10-28","last_updated":"2025-10-28","open_weights":true,"status":"deprecated"},
  "nvidia.nemotron-nano-3-30b": {"name":"NVIDIA Nemotron Nano 3 30B","description":"Small Nemotron 3 MoE for efficient coding, math, and long-context agents","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-nano-30b-a3b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":8192},"cost":{"input":0.06,"output":0.24},"knowledge":"2025-06-25","release_date":"2025-12-15","last_updated":"2025-12-23","open_weights":true},
  "nvidia.nemotron-nano-9b-v2": {"name":"NVIDIA Nemotron Nano 9B v2","description":"Compact Nemotron model for efficient reasoning and deployable AI agents","family":"nemotron","canonical_model_id":"nvidia/nemotron-nano-9b-v2","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":0.06,"output":0.23},"knowledge":"2024-09","release_date":"2025-08-18","last_updated":"2025-08-18","open_weights":true,"status":"deprecated"},
  "nvidia.nemotron-super-3-120b": {"name":"NVIDIA Nemotron 3 Super 120B A12B","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-super-120b-a12b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.15,"output":0.65},"knowledge":"2025-06","release_date":"2026-03-11","last_updated":"2026-03-11","open_weights":true},
  "openai.gpt-5.4": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","canonical_model_id":"openai/gpt-5.4","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.75,"output":16.5,"cache_read":0.275},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-06-01","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-5.5": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","canonical_model_id":"openai/gpt-5.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":33,"cache_read":0.55},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-06-01","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.22,"output":1.32,"cache_read":0.022,"cache_write":0.275,"tiers":[{"input":0.44,"output":1.98,"cache_read":0.044,"cache_write":0.55,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.44,"output":1.98,"cache_read":0.044,"cache_write":0.55}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-5.6-sol": {"name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4.4,"output":22,"cache_read":0.44,"cache_write":5.5,"tiers":[{"input":8.8,"output":33,"cache_read":0.88,"cache_write":11,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8.8,"output":33,"cache_read":0.88,"cache_write":11}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-5.6-terra": {"name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.2,"output":13.2,"cache_read":0.22,"cache_write":2.75,"tiers":[{"input":4.4,"output":19.8,"cache_read":0.44,"cache_write":5.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4.4,"output":19.8,"cache_read":0.44,"cache_write":5.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":11,"output":55,"cache_read":1.1,"cache_write":13.75,"tiers":[{"input":22,"output":82.5,"cache_read":2.2,"cache_write":27.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":22,"output":82.5,"cache_read":2.2,"cache_write":27.5}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"},"experimental":{"modes":{"ultrafast":{"cost":{"input":66,"output":330,"cache_read":6.6,"cache_write":82.5},"provider":{"body":{"service_tier":"ultrafast"}}}}}},
  "openai.gpt-6-luna": {"name":"GPT-6 Luna","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.11,"output":0.55,"cache_read":0.011,"cache_write":0.1375,"tiers":[{"input":0.22,"output":0.825,"cache_read":0.022,"cache_write":0.275,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.22,"output":0.825,"cache_read":0.022,"cache_write":0.275}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-6-sol": {"name":"GPT-6 Sol","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.22,"cache_write":2.75,"tiers":[{"input":4.4,"output":16.5,"cache_read":0.44,"cache_write":5.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4.4,"output":16.5,"cache_read":0.44,"cache_write":5.5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-6.1-sol": {"name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.11,"cache_write":2.75,"tiers":[{"input":4.4,"output":16.5,"cache_read":0.22,"cache_write":5.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4.4,"output":16.5,"cache_read":0.22,"cache_write":5.5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"},"experimental":{"modes":{"ultrafast":{"cost":{"input":13.2,"output":66,"cache_read":0.66,"cache_write":16.5},"provider":{"body":{"service_tier":"ultrafast"}}}}}},
  "openai.gpt-6.1-terra": {"name":"GPT-6.1 Terra","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1.1,"output":5.5,"cache_read":0.055,"cache_write":1.375,"tiers":[{"input":2.2,"output":8.25,"cache_read":0.11,"cache_write":2.75,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2.2,"output":8.25,"cache_read":0.11,"cache_write":2.75}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "openai.gpt-oss-120b": {"name":"gpt-oss-120b","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.15,"output":0.6},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/v1","shape":"responses"}},
  "openai.gpt-oss-120b-1:0": {"name":"gpt-oss-120b","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":128000},"cost":{"input":0.15,"output":0.6},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "openai.gpt-oss-20b": {"name":"gpt-oss-20b","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.07,"output":0.3},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/v1","shape":"responses"}},
  "openai.gpt-oss-20b-1:0": {"name":"gpt-oss-20b","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":128000},"cost":{"input":0.07,"output":0.3},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "openai.gpt-oss-safeguard-120b": {"name":"GPT OSS Safeguard 120B","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-safeguard-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6},"release_date":"2025-10-29","last_updated":"2025-10-29","open_weights":true},
  "openai.gpt-oss-safeguard-20b": {"name":"GPT OSS Safeguard 20B","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-safeguard-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.07,"output":0.2},"release_date":"2025-10-29","last_updated":"2025-10-29","open_weights":true},
  "qwen.qwen3-235b-a22b-2507-v1:0": {"name":"Qwen3 235B-A22B Instruct 2507","description":"Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-instruct-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.22,"output":0.88},"release_date":"2025-07-21","last_updated":"2025-09-18","open_weights":true},
  "qwen.qwen3-32b-v1:0": {"name":"Qwen3 32B","description":"Dense open Qwen model for self-hosted chat, reasoning, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3-32b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":16384},"cost":{"input":0.15,"output":0.6},"knowledge":"2025-04","release_date":"2025-04-28","last_updated":"2025-09-18","open_weights":true},
  "qwen.qwen3-coder-30b-a3b-v1:0": {"name":"Qwen3-Coder 30B-A3B Instruct","description":"Smaller Qwen coder for efficient local agents and repo-level fixes","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-30b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.15,"output":0.6},"knowledge":"2025-04","release_date":"2025-07-31","last_updated":"2025-09-18","open_weights":true,"status":"deprecated"},
  "qwen.qwen3-coder-480b-a35b-v1:0": {"name":"Qwen3-Coder 480B-A35B Instruct","description":"Open Qwen coding heavyweight for repository reasoning and agentic engineering","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-480b-a35b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.45,"output":1.8},"knowledge":"2025-04","release_date":"2025-07-23","last_updated":"2025-09-18","open_weights":true,"status":"deprecated"},
  "qwen.qwen3-coder-next": {"name":"Qwen3 Coder Next","description":"Open-weight Qwen coding model for agents, repository edits, and multi-turn tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-next","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.5,"output":1.2},"knowledge":"2025-09","release_date":"2026-02-03","last_updated":"2026-02-06","open_weights":true},
  "qwen.qwen3-next-80b-a3b": {"name":"Qwen3-Next 80B-A3B Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-next-80b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262000},"cost":{"input":0.15,"output":1.2},"knowledge":"2025-04","release_date":"2025-09-11","last_updated":"2025-11-25","open_weights":true},
  "qwen.qwen3-vl-235b-a22b": {"name":"Qwen3 VL 235B A22B Instruct","description":"Qwen vision-language instruct model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3-vl-235b-a22b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262000},"cost":{"input":0.53,"output":2.66},"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-11-25","open_weights":true},
  "stability.sd3-5-large-v1:0": {"name":"Stable Diffusion 3.5 Large","modalities":{"input":["text","image"]},"rig":{"pinned":["name","modalities"]}},
  "stability.stable-image-core-v1:1": {"name":"Stable Image Core","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "stability.stable-image-ultra-v1:1": {"name":"Stable Image Ultra","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "us-gov.openai.gpt-oss-120b-1:0": {"name":"gpt-oss-120b (GovCloud)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.18,"output":0.72},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "us-gov.openai.gpt-oss-20b-1:0": {"name":"gpt-oss-20b (GovCloud)","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.084,"output":0.36},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "us.amazon.nova-2-lite-v1:0": {"name":"Nova 2 Lite (US)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova","canonical_model_id":"amazon/nova-2-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":65535},"cost":{"input":0.33,"output":2.75,"cache_read":0.0825,"cache_write":0.33},"knowledge":"2025-10","release_date":"2025-12-02","last_updated":"2025-12-01","open_weights":false},
  "us.amazon.nova-lite-v1:0": {"name":"Nova Lite (US)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-lite","canonical_model_id":"amazon/nova-lite","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.06,"output":0.24,"cache_read":0.015,"cache_write":0.06},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "us.amazon.nova-micro-v1:0": {"name":"Nova Micro (US)","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-micro","canonical_model_id":"amazon/nova-micro","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":10000},"cost":{"input":0.035,"output":0.14,"cache_read":0.00875,"cache_write":0.035},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "us.amazon.nova-premier-v1:0": {"name":"Nova Premier (US)","description":"Multimodal model for complex analysis, long-context understanding, tool use, and model distillation","family":"nova","canonical_model_id":"amazon/nova-premier","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":10000},"cost":{"input":2.5,"output":12.5,"cache_read":0.625,"cache_write":2.5},"knowledge":"2024-10","release_date":"2025-04-30","last_updated":"2025-04-30","open_weights":false,"status":"deprecated"},
  "us.amazon.nova-pro-v1:0": {"name":"Nova Pro (US)","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova-pro","canonical_model_id":"amazon/nova-pro","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":300000,"output":10000},"cost":{"input":0.8,"output":3.2,"cache_read":0.2,"cache_write":0.8},"knowledge":"2024-10","release_date":"2024-12-03","last_updated":"2024-12-03","open_weights":false},
  "us.anthropic.claude-fable-5": {"name":"Claude Fable 5 (US)","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":11,"output":55,"cache_read":1.1,"cache_write":13.75},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "us.anthropic.claude-fable-5-1": {"name":"Claude Fable 5.1 (US)","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":11,"output":55,"cache_read":0.275,"cache_write":13.75},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never"}},
  "us.anthropic.claude-haiku-4-5-20251001-v1:0": {"name":"Claude Haiku 4.5 (US)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1.1,"output":5.5,"cache_read":0.11,"cache_write":1.375},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "us.anthropic.claude-haiku-5-5": {"name":"Claude Haiku 5.5 (US)","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.11,"output":0.55,"cache_read":0.011,"cache_write":0.1375,"tiers":[{"input":0.55,"output":2.75,"cache_read":0.055,"cache_write":0.6875,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "us.anthropic.claude-opus-4-1-20250805-v1:0": {"name":"Claude Opus 4.1 (US)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-1-20250805","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "us.anthropic.claude-opus-4-5-20251101-v1:0": {"name":"Claude Opus 4.5 (US)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5-20251101","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "us.anthropic.claude-opus-4-6-v1": {"name":"Claude Opus 4.6 (US)","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "us.anthropic.claude-opus-4-7": {"name":"Claude Opus 4.7 (US)","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "us.anthropic.claude-opus-4-8": {"name":"Claude Opus 4.8 (US)","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "us.anthropic.claude-opus-5": {"name":"Claude Opus 5 (US)","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5.5,"output":27.5,"cache_read":0.55,"cache_write":6.875},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "us.anthropic.claude-opus-5-5": {"name":"Claude Opus 5.5 (US)","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4.4,"output":22,"cache_read":0.22,"cache_write":5.5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "us.anthropic.claude-sonnet-4-20250514-v1:0": {"name":"Claude Sonnet 4 (US)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-20250514","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "us.anthropic.claude-sonnet-4-5-20250929-v1:0": {"name":"Claude Sonnet 4.5 (US)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5-20250929","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"status":"deprecated","rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "us.anthropic.claude-sonnet-4-6": {"name":"Claude Sonnet 4.6 (US)","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "us.anthropic.claude-sonnet-5": {"name":"Claude Sonnet 5 (US)","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.22,"cache_write":2.75},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "us.anthropic.claude-sonnet-5-5": {"name":"Claude Sonnet 5.5 (US)","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.11,"cache_write":2.75},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "us.deepseek.r1-v1:0": {"name":"DeepSeek-R1 (US)","description":"Classic open reasoning model for transparent math, coding, and deliberate problem solving","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-r1","reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":1.35,"output":5.4},"knowledge":"2024-07","release_date":"2025-01-20","last_updated":"2025-05-29","open_weights":true,"status":"deprecated"},
  "us.meta.llama3-1-70b-instruct-v1:0": {"name":"Llama 3.1 70B Instruct (US)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.1-70b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.72,"output":0.72},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true,"status":"deprecated"},
  "us.meta.llama3-1-8b-instruct-v1:0": {"name":"Llama 3.1 8B Instruct (US)","description":"Compact open Llama model for lightweight chat, drafting, and self-hosting","family":"llama","canonical_model_id":"meta/llama-3.1-8b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.22,"output":0.22},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true,"status":"deprecated"},
  "us.meta.llama3-3-70b-instruct-v1:0": {"name":"Llama 3.3 70B Instruct (US)","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","canonical_model_id":"meta/llama-3.3-70b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.72,"output":0.72},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true,"status":"deprecated"},
  "us.meta.llama4-maverick-17b-instruct-v1:0": {"name":"Llama 4 Maverick 17B Instruct (US)","description":"Open multimodal Llama for strong reasoning with efficient everyday serving","family":"llama","canonical_model_id":"meta/llama-4-maverick-17b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":8192},"cost":{"input":0.24,"output":0.97},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true,"status":"deprecated"},
  "us.meta.llama4-scout-17b-instruct-v1:0": {"name":"Llama 4 Scout 17B Instruct (US)","description":"Open Llama with long-context vision for efficient multimodal agents","family":"llama","canonical_model_id":"meta/llama-4-scout-17b-instruct","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":10000000,"output":8192},"cost":{"input":0.17,"output":0.66},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true,"status":"deprecated"},
  "us.mistral.pixtral-large-2502-v1:0": {"name":"Pixtral Large (25.02) (US)","description":"Mistral vision-language model for image understanding and multimodal chat","family":"pixtral","canonical_model_id":"mistral/pixtral-large-2502","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":2,"output":6},"release_date":"2025-04-08","last_updated":"2025-04-08","open_weights":false},
  "us.moonshotai.kimi-k3": {"name":"Kimi K3 (US)","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":128000},"cost":{"input":3.3,"output":16.5,"cache_read":0.33,"cache_write":4.125},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "us.openai.gpt-5.6-luna": {"name":"GPT-5.6 Luna (US)","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.22,"output":1.32,"cache_read":0.022,"cache_write":0.275,"tiers":[{"input":0.44,"output":1.98,"cache_read":0.044,"cache_write":0.55,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.44,"output":1.98,"cache_read":0.044,"cache_write":0.55}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "us.openai.gpt-5.6-sol": {"name":"GPT-5.6 Sol (US)","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4.4,"output":22,"cache_read":0.44,"cache_write":5.5,"tiers":[{"input":8.8,"output":33,"cache_read":0.88,"cache_write":11,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8.8,"output":33,"cache_read":0.88,"cache_write":11}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "us.openai.gpt-5.6-terra": {"name":"GPT-5.6 Terra (US)","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.2,"output":13.2,"cache_read":0.22,"cache_write":2.75,"tiers":[{"input":4.4,"output":19.8,"cache_read":0.44,"cache_write":5.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4.4,"output":19.8,"cache_read":0.44,"cache_write":5.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "us.openai.gpt-6-astra": {"name":"GPT-6 Astra (US)","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":11,"output":55,"cache_read":1.1,"cache_write":13.75,"tiers":[{"input":22,"output":82.5,"cache_read":2.2,"cache_write":27.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":22,"output":82.5,"cache_read":2.2,"cache_write":27.5}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"experimental":{"modes":{"ultrafast":{"cost":{"input":66,"output":330,"cache_read":6.6,"cache_write":82.5},"provider":{"body":{"additionalModelRequestFields":{"service_tier":"ultrafast"}}}}}}},
  "us.openai.gpt-6-luna": {"name":"GPT-6 Luna (US)","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.11,"output":0.55,"cache_read":0.011,"cache_write":0.1375,"tiers":[{"input":0.22,"output":0.825,"cache_read":0.022,"cache_write":0.275,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.22,"output":0.825,"cache_read":0.022,"cache_write":0.275}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "us.openai.gpt-6-sol": {"name":"GPT-6 Sol (US)","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.22,"cache_write":2.75,"tiers":[{"input":4.4,"output":16.5,"cache_read":0.44,"cache_write":5.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4.4,"output":16.5,"cache_read":0.44,"cache_write":5.5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "us.openai.gpt-6.1-sol": {"name":"GPT-6.1 Sol (US)","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.2,"output":11,"cache_read":0.11,"cache_write":2.75,"tiers":[{"input":4.4,"output":16.5,"cache_read":0.22,"cache_write":5.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4.4,"output":16.5,"cache_read":0.22,"cache_write":5.5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"experimental":{"modes":{"ultrafast":{"cost":{"input":13.2,"output":66,"cache_read":0.66,"cache_write":16.5},"provider":{"body":{"additionalModelRequestFields":{"service_tier":"ultrafast"}}}}}}},
  "us.openai.gpt-6.1-terra": {"name":"GPT-6.1 Terra (US)","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1.1,"output":5.5,"cache_read":0.055,"cache_write":1.375,"tiers":[{"input":2.2,"output":8.25,"cache_read":0.11,"cache_write":2.75,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2.2,"output":8.25,"cache_read":0.11,"cache_write":2.75}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false},
  "us.writer.palmyra-x4-v1:0": {"name":"Palmyra X4 (US)","description":"Enterprise language model for workflow automation, coding, data analysis, and tool use","family":"palmyra","canonical_model_id":"writer/palmyra-x4","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":122880,"output":8192},"cost":{"input":2.5,"output":10},"release_date":"2024-10-09","last_updated":"2025-04-28","open_weights":false,"status":"deprecated"},
  "us.writer.palmyra-x5-v1:0": {"name":"Palmyra X5 (US)","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"palmyra","canonical_model_id":"writer/palmyra-x5","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1040000,"input":1040000,"output":8192},"cost":{"input":0.6,"output":6},"release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":false,"status":"deprecated"},
  "us.xai.grok-4.6": {"name":"Grok 4.6 (US)","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2.2,"output":6.6,"cache_read":0.55},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-18","open_weights":false},
  "us.xai.grok-4.7": {"name":"Grok 4.7 (US)","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2.2,"output":6.6,"cache_read":0.55},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-28","open_weights":false},
  "us.zai.glm-5.3": {"name":"GLM-5.3 (US)","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":1.848,"output":5.808,"cache_read":0.3432,"cache_write":2.31},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "writer.palmyra-x4-v1:0": {"name":"Palmyra X4","description":"Enterprise language model for workflow automation, coding, data analysis, and tool use","family":"palmyra","canonical_model_id":"writer/palmyra-x4","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":122880,"output":8192},"cost":{"input":2.5,"output":10},"release_date":"2024-10-09","last_updated":"2025-04-28","open_weights":false,"status":"deprecated"},
  "writer.palmyra-x5-v1:0": {"name":"Palmyra X5","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"palmyra","canonical_model_id":"writer/palmyra-x5","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1040000,"input":1040000,"output":8192},"cost":{"input":0.6,"output":6},"release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":false,"status":"deprecated"},
  "xai.grok-4.3": {"name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.25,"output":2.5,"cache_read":0.2},"release_date":"2026-04-17","last_updated":"2026-06-28","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "xai.grok-4.6": {"name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2.2,"output":6.6,"cache_read":0.55},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-18","open_weights":false,"provider":{"npm":"@ai-sdk/amazon-bedrock/mantle","api":"https://bedrock-mantle.${AWS_REGION}.api.aws/openai/v1","shape":"responses"}},
  "zai.glm-4.7": {"name":"GLM-4.7","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.7","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":0.6,"output":2.2},"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","open_weights":true,"status":"deprecated"},
  "zai.glm-4.7-flash": {"name":"GLM-4.7-Flash","description":"Budget GLM lane for fast coding help, routing, and everyday automation","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.7-flash","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":0.07,"output":0.4},"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","open_weights":true,"status":"deprecated"},
  "zai.glm-5": {"name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":1,"output":3.2},"release_date":"2026-02-12","last_updated":"2026-03-18","open_weights":true}
}},
"azure.openai": {"models": {
  "claude-fable-5": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"status":"beta","provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-fable-5-1": {"name":"Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-haiku-4-5": {"name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-11-18","last_updated":"2025-11-18","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-haiku-5-5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-mythos-5": {"name":"Claude Mythos 5","description":"Restricted Claude model for advanced cybersecurity and biology research workflows","family":"claude-mythos","canonical_model_id":"anthropic/claude-mythos-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"status":"beta","provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-opus-4-5": {"name":"Claude Opus 4.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-03-31","release_date":"2025-11-24","last_updated":"2025-08-01","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-opus-4-6": {"name":"Claude Opus 4.6","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-07-17","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-opus-4-7": {"name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-opus-4-8": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}},"knowledge":"2025-12-31","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-opus-5": {"name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-opus-5-5": {"name":"Claude Opus 5.5","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-sonnet-4-5": {"name":"Claude Sonnet 4.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-07-31","release_date":"2025-11-18","last_updated":"2025-11-18","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-sonnet-4-6": {"name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-sonnet-5": {"name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "claude-sonnet-5-5": {"name":"Claude Sonnet 5.5","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/anthropic/v1"}},
  "codestral-2501": {"name":"Codestral 25.01","description":"Mistral coding model for code completion, generation, and developer workflows","family":"codestral","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":256000},"cost":{"input":0.3,"output":0.9},"knowledge":"2024-03","release_date":"2025-01-01","last_updated":"2025-01-01","open_weights":false,"provider":{"shape":"completions"}},
  "codex-mini": {"name":"Codex Mini","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":1.5,"output":6,"cache_read":0.375},"knowledge":"2024-04","release_date":"2025-05-16","last_updated":"2025-05-16","open_weights":false,"status":"deprecated"},
  "cohere-command-a": {"name":"Command A","description":"Cohere command model for multilingual enterprise agents, tools, and chat","family":"command-a","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":2.5,"output":10},"knowledge":"2024-06-01","release_date":"2025-03-13","last_updated":"2025-03-13","open_weights":true,"provider":{"shape":"completions"}},
  "cohere-embed-v-4-0": {"name":"Embed v4","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"cohere-embed","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":1536},"cost":{"input":0.12,"output":0},"release_date":"2025-04-15","last_updated":"2025-04-15","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "cohere-embed-v3-english": {"name":"Embed v3 English","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"cohere-embed","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":512,"output":1024},"cost":{"input":0.1,"output":0},"release_date":"2023-11-07","last_updated":"2023-11-07","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "cohere-embed-v3-multilingual": {"name":"Embed v3 Multilingual","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"cohere-embed","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":512,"output":1024},"cost":{"input":0.1,"output":0},"release_date":"2023-11-07","last_updated":"2023-11-07","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "deepseek-v3.2": {"name":"DeepSeek-V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":128000},"cost":{"input":0.58,"output":1.68},"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","open_weights":true,"provider":{"shape":"completions"}},
  "deepseek-v3.2-speciale": {"name":"DeepSeek-V3.2-Speciale","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":128000},"cost":{"input":0.58,"output":1.68},"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","open_weights":true,"provider":{"shape":"completions"}},
  "deepseek-v4-flash": {"name":"DeepSeek-V4-Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":0.19,"output":0.51},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "deepseek-v4-pro": {"name":"DeepSeek-V4-Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":1.74,"output":3.48},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "deepseek-v4.1-flash": {"name":"DeepSeek-V4.1-Flash","description":"DeepSeek V4.1 Flash model for reasoning and agentic coding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.3,"output":1.2,"cache_read":0.006},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true,"status":"beta","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "gpt-3.5-turbo": {"name":"GPT-3.5-turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":16385,"output":4096},"cost":{"input":0.5,"output":1.5,"cache_read":0},"knowledge":"2021-09-01","release_date":"2023-03-01","last_updated":"2023-11-06","open_weights":false,"status":"deprecated","rig":{"cache":["short"],"pinned":["name","description","family","reasoning","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","knowledge","release_date","last_updated","open_weights","status"]}},
  "gpt-3.5-turbo-0125": {"name":"GPT-3.5 Turbo 0125","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":16384,"output":4096},"cost":{"input":0.5,"output":1.5},"knowledge":"2021-08","release_date":"2024-01-25","last_updated":"2024-01-25","open_weights":false,"status":"deprecated","provider":{"shape":"completions"}},
  "gpt-3.5-turbo-1106": {"name":"GPT-3.5 Turbo 1106","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":16384,"output":4096},"cost":{"input":1,"output":2},"knowledge":"2021-08","release_date":"2023-11-06","last_updated":"2023-11-06","open_weights":false,"status":"deprecated","provider":{"shape":"completions"}},
  "gpt-3.5-turbo-16k": {"name":"gpt-3.5-turbo-16k","tool_call":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","tool_call","modalities"]}},
  "gpt-3.5-turbo-instruct": {"name":"GPT-3.5 Turbo Instruct","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":4096,"output":4096},"cost":{"input":1.5,"output":2},"knowledge":"2021-08","release_date":"2023-09-21","last_updated":"2023-09-21","open_weights":false,"status":"deprecated"},
  "gpt-4": {"name":"GPT-4","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":8192},"cost":{"input":30,"output":60},"knowledge":"2023-11","release_date":"2023-11-06","last_updated":"2024-04-09","open_weights":false,"status":"deprecated","rig":{"cache":["short"],"pinned":["name","description","family","reasoning","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","knowledge","release_date","last_updated","open_weights","status"]}},
  "gpt-4-32k": {"name":"GPT-4 32K","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":32768,"output":32768},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4-turbo": {"name":"GPT-4 Turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","canonical_model_id":"openai/gpt-4-turbo","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":10,"output":30},"knowledge":"2023-12","release_date":"2023-11-06","last_updated":"2024-04-09","open_weights":false,"status":"deprecated","provider":{"shape":"completions"}},
  "gpt-4-turbo-vision": {"name":"GPT-4 Turbo Vision","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":10,"output":30},"knowledge":"2023-11","release_date":"2023-11-06","last_updated":"2024-04-09","open_weights":false,"status":"deprecated","provider":{"shape":"completions"}},
  "gpt-4.1": {"name":"GPT-4.1","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","canonical_model_id":"openai/gpt-4.1","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false,"status":"deprecated"},
  "gpt-4.1-mini": {"name":"GPT-4.1 mini","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","canonical_model_id":"openai/gpt-4.1-mini","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false,"status":"deprecated"},
  "gpt-4.1-nano": {"name":"GPT-4.1 nano","description":"Tiny GPT-4.1 option for classification, routing, and very high-volume tasks","family":"gpt-nano","canonical_model_id":"openai/gpt-4.1-nano","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.025},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false,"status":"deprecated"},
  "gpt-4o": {"name":"GPT-4o","description":"Omni-era GPT for multimodal chat, practical coding, and general assistants","family":"gpt","canonical_model_id":"openai/gpt-4o","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","open_weights":false,"status":"deprecated"},
  "gpt-4o-mini": {"name":"GPT-4o mini","description":"Small omni GPT for cheap multimodal assistance and production-scale traffic","family":"gpt-mini","canonical_model_id":"openai/gpt-4o-mini","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075},"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","open_weights":false,"status":"deprecated"},
  "gpt-4o-realtime-preview": {"name":"gpt-4o-realtime-preview","tool_call":true,"modalities":{"input":["text","audio"]},"rig":{"pinned":["name","tool_call","modalities"]}},
  "gpt-5": {"name":"GPT-5","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false},
  "gpt-5-codex": {"name":"GPT-5-Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-09-15","last_updated":"2025-09-15","open_weights":false},
  "gpt-5-mini": {"name":"GPT-5 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false},
  "gpt-5-nano": {"name":"GPT-5 Nano","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false},
  "gpt-5-pro": {"name":"GPT-5 Pro","description":"Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning","family":"gpt-pro","canonical_model_id":"openai/gpt-5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":15,"output":120},"knowledge":"2024-09-30","release_date":"2025-10-06","last_updated":"2025-10-06","open_weights":false,"rig":{"sampling":"never"}},
  "gpt-5.1": {"name":"GPT-5.1","description":"Sharper GPT-5 generation for coding, product work, and tool-assisted tasks","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-11-14","last_updated":"2025-11-14","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":2.5,"output":20,"cache_read":0.25},"provider":{"body":{"service_tier":"priority"}}}}}},
  "gpt-5.1-codex": {"name":"GPT-5.1 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-11-14","last_updated":"2025-11-14","open_weights":false},
  "gpt-5.1-codex-max": {"name":"GPT-5.1 Codex Max","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.1-codex-max","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false},
  "gpt-5.1-codex-mini": {"name":"GPT-5.1 Codex Mini","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025},"knowledge":"2024-09-30","release_date":"2025-11-14","last_updated":"2025-11-14","open_weights":false},
  "gpt-5.2": {"name":"GPT-5.2","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":3.5,"output":28,"cache_read":0.35},"provider":{"body":{"service_tier":"priority"}}}}}},
  "gpt-5.2-codex": {"name":"GPT-5.2 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-01-14","last_updated":"2026-01-14","open_weights":false},
  "gpt-5.3-codex": {"name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-24","last_updated":"2026-02-24","open_weights":false},
  "gpt-5.4": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","canonical_model_id":"openai/gpt-5.4","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":5,"output":30,"cache_read":0.5},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-mini": {"name":"GPT-5.4 Mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","canonical_model_id":"openai/gpt-5.4-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":1.5,"output":9,"cache_read":0.15},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-nano": {"name":"GPT-5.4 Nano","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","canonical_model_id":"openai/gpt-5.4-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-pro": {"name":"GPT-5.4 Pro","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","canonical_model_id":"openai/gpt-5.4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"rig":{"sampling":"reasoning_off"}},
  "gpt-5.5": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","canonical_model_id":"openai/gpt-5.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"knowledge":"2025-12-01","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":12.5,"output":75,"cache_read":1.25},"provider":{"body":{"service_tier":"priority"}}}}}},
  "gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-sol": {"name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":8,"output":40,"cache_read":0.8,"cache_write":10},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-terra": {"name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4,"output":24,"cache_read":0.4,"cache_write":5},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-luna": {"name":"GPT-6 Luna","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-sol": {"name":"GPT-6 Sol","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-sol": {"name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-terra": {"name":"GPT-6.1 Terra","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false,"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-chat-latest": {"name":"GPT Chat Latest","description":"Compact GPT model for low-latency assistance and high-volume workloads","canonical_model_id":"openai/gpt-5.5-instant","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5},"knowledge":"2025-12-01","release_date":"2026-05-05","last_updated":"2026-05-28","open_weights":false,"status":"beta"},
  "gpt-image-1": {"name":"GPT-Image-1","description":"OpenAI image model for production generation, edits, and brand-safe visual workflows","family":"gpt-image","type":"image","canonical_model_id":"openai/gpt-image-1","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["image"]},"limit":{"context":0,"output":0},"cost":{"input":5,"output":40,"cache_read":1.25},"release_date":"2025-04-24","last_updated":"2025-04-24","open_weights":false},
  "gpt-image-1.5": {"name":"GPT-Image-1.5","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","canonical_model_id":"openai/gpt-image-1.5","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":0,"output":0},"cost":{"input":5,"output":32,"cache_read":1.25},"release_date":"2025-11-25","last_updated":"2025-11-25","open_weights":false},
  "gpt-image-2": {"name":"GPT-Image-2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","canonical_model_id":"openai/gpt-image-2","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["image"]},"limit":{"context":0,"output":0},"cost":{"input":5,"output":30,"cache_read":1.25},"release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false},
  "gpt-image-2.5-flare": {"name":"GPT Image 2.5 Flare","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","canonical_model_id":"openai/gpt-image-2.5-flare","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["image"]},"limit":{"context":0,"output":0},"cost":{"input":5,"output":30,"cache_read":1.25},"release_date":"2026-09-08","last_updated":"2026-09-08","open_weights":false},
  "gpt-image-2.5-sunburst": {"name":"GPT Image 2.5 Sunburst","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","canonical_model_id":"openai/gpt-image-2.5-sunburst","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["image"]},"limit":{"context":0,"output":0},"cost":{"input":5,"output":30,"cache_read":1.25},"release_date":"2026-09-08","last_updated":"2026-09-08","open_weights":false},
  "grok-4-1-fast-non-reasoning": {"name":"Grok 4.1 Fast (Non-Reasoning)","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","family":"grok","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.2,"output":0.5,"cache_read":0.05},"release_date":"2025-06-27","last_updated":"2025-06-27","open_weights":false,"provider":{"shape":"completions"}},
  "grok-4-1-fast-reasoning": {"name":"Grok 4.1 Fast (Reasoning)","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","family":"grok","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"input":128000,"output":8192},"cost":{"input":0.2,"output":0.5,"cache_read":0.05},"release_date":"2025-06-27","last_updated":"2025-06-27","open_weights":false,"provider":{"shape":"completions"}},
  "grok-4-20-non-reasoning": {"name":"Grok 4.20 (Non-Reasoning)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262000,"output":8192},"cost":{"input":2,"output":6},"knowledge":"2025-09","release_date":"2026-04-08","last_updated":"2026-04-08","open_weights":false,"status":"beta","provider":{"shape":"completions"}},
  "grok-4-20-reasoning": {"name":"Grok 4.20 (Reasoning)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262000,"output":8192},"cost":{"input":2,"output":6},"knowledge":"2025-09","release_date":"2026-04-08","last_updated":"2026-04-08","open_weights":false,"status":"beta","provider":{"shape":"completions"}},
  "grok-4.6": {"name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":128000},"cost":{"input":2,"output":6,"cache_read":0.5},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":false,"status":"beta","provider":{"shape":"completions"}},
  "grok-4.7": {"name":"Grok 4.7","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false,"status":"beta","provider":{"shape":"completions"}},
  "kimi-k2.5": {"name":"Kimi K2.5","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.6,"output":3},"knowledge":"2025-01","release_date":"2026-01-27","last_updated":"2026-01-27","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "kimi-k2.6": {"name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.6","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4},"knowledge":"2025-01","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "kimi-k2.7-code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${AZURE_RESOURCE_NAME}.services.ai.azure.com/models"}},
  "llama-3.3-70b-instruct": {"name":"Llama-3.3-70B-Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":0.71,"output":0.71},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true,"provider":{"shape":"completions"}},
  "llama-4-maverick-17b-128e-instruct-fp8": {"name":"Llama 4 Maverick 17B 128E Instruct FP8","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":16384},"cost":{"input":0.25,"output":1},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true,"provider":{"shape":"completions"}},
  "llama-4-scout-17b-16e-instruct": {"name":"Llama 4 Scout 17B 16E Instruct","description":"Open multimodal Llama model for long-context analysis and efficient agents","family":"llama","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.2,"output":0.78},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true,"provider":{"shape":"completions"}},
  "ministral-3b": {"name":"Ministral 3B","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.04,"output":0.04},"knowledge":"2024-03","release_date":"2024-10-22","last_updated":"2024-10-22","open_weights":true,"provider":{"shape":"completions"}},
  "mistral-medium-2505": {"name":"Mistral Medium 3","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":128000},"cost":{"input":0.4,"output":2},"knowledge":"2025-05","release_date":"2025-05-07","last_updated":"2025-05-07","open_weights":false,"provider":{"shape":"completions"}},
  "mistral-small-2503": {"name":"Mistral Small 3.1","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":0.1,"output":0.3},"knowledge":"2024-09","release_date":"2025-03-01","last_updated":"2025-03-01","open_weights":false,"provider":{"shape":"completions"}},
  "model-router": {"name":"Model Router","description":"Automatic model router for matching prompts to suitable backends and budgets","family":"model-router","reasoning":false,"tool_call":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":16384},"cost":{"input":0.14,"output":0},"release_date":"2025-05-19","last_updated":"2025-11-18","open_weights":false,"provider":{"npm":"@ai-sdk/azure","shape":"completions"}},
  "o1": {"name":"o1","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5},"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","open_weights":false,"status":"deprecated"},
  "o1-mini": {"name":"o1-mini","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":false,"modalities":{"input":["text"]},"limit":{"context":128000,"output":65536},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities","limit.context","limit.output","status"],"reasoning_control":"none","sampling":"never"}},
  "o1-preview": {"name":"o1-preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":false,"modalities":{"input":["text"]},"limit":{"context":128000,"output":32768},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities","limit.context","limit.output","status"],"reasoning_control":"none","sampling":"never"}},
  "o3": {"name":"o3","description":"Deliberate o-series reasoner for hard math, coding, and multi-step analysis","family":"o","canonical_model_id":"openai/o3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":0.5},"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","open_weights":false,"rig":{"sampling":"never"}},
  "o3-mini": {"name":"o3-mini","description":"Smaller o-series reasoner for economical coding, math, and planning tasks","family":"o-mini","canonical_model_id":"openai/o3-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55},"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","open_weights":false,"status":"deprecated","rig":{"sampling":"never"}},
  "o4-mini": {"name":"o4-mini","description":"Fast o-series model for compact reasoning, coding, and tool use","family":"o-mini","canonical_model_id":"openai/o4-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.275},"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","open_weights":false,"status":"deprecated","rig":{"sampling":"never"}},
  "phi-4": {"name":"Phi-4","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","family":"phi","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.125,"output":0.5},"knowledge":"2023-10","release_date":"2024-12-11","last_updated":"2024-12-11","open_weights":true,"provider":{"shape":"completions"}},
  "phi-4-mini": {"name":"Phi-4-mini","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"phi","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.075,"output":0.3},"knowledge":"2023-10","release_date":"2024-12-11","last_updated":"2024-12-11","open_weights":true,"provider":{"shape":"completions"}},
  "phi-4-mini-reasoning": {"name":"Phi-4-mini-reasoning","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"phi","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.075,"output":0.3},"knowledge":"2023-10","release_date":"2024-12-11","last_updated":"2024-12-11","open_weights":true,"provider":{"shape":"completions"}},
  "phi-4-multimodal": {"name":"Phi-4-multimodal","description":"Multimodal model for analyzing text, images, documents, and rich media","family":"phi","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.08,"output":0.32,"input_audio":4},"knowledge":"2023-10","release_date":"2024-12-11","last_updated":"2024-12-11","open_weights":true,"provider":{"shape":"completions"}},
  "phi-4-reasoning": {"name":"Phi-4-reasoning","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"phi","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32000,"output":4096},"cost":{"input":0.125,"output":0.5},"knowledge":"2023-10","release_date":"2024-12-11","last_updated":"2024-12-11","open_weights":true,"provider":{"shape":"completions"}},
  "phi-4-reasoning-plus": {"name":"Phi-4-reasoning-plus","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"phi","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32000,"output":4096},"cost":{"input":0.125,"output":0.5},"knowledge":"2023-10","release_date":"2024-12-11","last_updated":"2024-12-11","open_weights":true,"provider":{"shape":"completions"}},
  "text-embedding-3-large": {"name":"text-embedding-3-large","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","type":"embedding","reasoning":false,"tool_call":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":3072},"cost":{"input":0.13,"output":0},"release_date":"2024-01-25","last_updated":"2024-01-25","open_weights":false},
  "text-embedding-3-small": {"name":"text-embedding-3-small","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","type":"embedding","reasoning":false,"tool_call":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":1536},"cost":{"input":0.02,"output":0},"release_date":"2024-01-25","last_updated":"2024-01-25","open_weights":false},
  "text-embedding-ada-002": {"name":"text-embedding-ada-002","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","type":"embedding","reasoning":false,"tool_call":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":1536},"cost":{"input":0.1,"output":0},"release_date":"2022-12-15","last_updated":"2022-12-15","open_weights":false}
}},
"chatgpt": {"models": {
  "gpt-5.3-chat-latest": {"name":"GPT-5.3 Chat (latest)","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":16384},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null},"status":"deprecated","rig":{"cache":["long"],"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","status"],"sampling":"reasoning_off"}},
  "gpt-5.3-codex": {"name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":3.5,"output":28,"cache_read":0.35},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","knowledge","release_date","last_updated","open_weights","experimental"],"sampling":"reasoning_off"}},
  "gpt-5.3-codex-spark": {"name":"GPT-5.3 Codex Spark","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex-spark","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"input":100000,"output":32000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","knowledge","release_date","last_updated","open_weights"],"sampling":"reasoning_off"}},
  "gpt-5.3-instant": {"name":"GPT-5.3 Instant","reasoning":false,"tool_call":true,"modalities":{"input":["text","image"]},"rig":{"cache":["long"],"pinned":["name","reasoning","tool_call","modalities"]}},
  "gpt-5.4": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5},"cache_write":null},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":5,"output":30,"cache_read":0.5},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"sampling":"reasoning_off"}},
  "gpt-5.4-pro": {"name":"GPT-5.4 Pro","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270},"cache_read":null,"cache_write":null},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights"],"sampling":"reasoning_off"}},
  "gpt-5.5": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1},"cache_write":null},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":12.5,"output":75,"cache_read":1.25},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"]}},
  "gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":0.4,"output":2.4,"cache_read":0.04,"cache_write":0.5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-sol": {"name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":8,"output":40,"cache_read":0.8,"cache_write":10},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-terra": {"name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4,"output":24,"cache_read":0.4,"cache_write":5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["long"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":20,"output":100,"cache_read":2,"cache_write":25},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}},"ultrafast":{"cost":{"input":60,"output":300,"cache_read":6,"cache_write":75},"provider":{"body":{"service_tier":"ultrafast"}}}}},"rig":{"cache":["long"],"chat_tools":"never","pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-luna": {"name":"GPT-6 Luna","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":0.2,"output":1,"cache_read":0.02,"cache_write":0.25},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["long"],"chat_tools":"reasoning_off","pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-sol": {"name":"GPT-6 Sol","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["long"],"chat_tools":"reasoning_off","pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-sol": {"name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}},"ultrafast":{"cost":{"input":12,"output":60,"cache_read":0.6,"cache_write":15},"provider":{"body":{"service_tier":"ultrafast"}}}}},"rig":{"cache":["long"],"chat_tools":"never","pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-terra": {"name":"GPT-6.1 Terra","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":null,"output":null,"cache_read":null,"cache_write":null,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["long"],"chat_tools":"never","pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.input","limit.output","cost.input","cost.output","cost.cache_read","cost.cache_write","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","experimental"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}}
}},
"cohere": {"models": {
  "c4ai-aya-expanse-32b": {"name":"Aya Expanse 32B","description":"Open multilingual model optimized for generation across 23 languages","canonical_model_id":"cohere/c4ai-aya-expanse-32b","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":0.5,"output":1.5},"release_date":"2024-10-24","last_updated":"2024-10-24","open_weights":true},
  "c4ai-aya-vision-32b": {"name":"Aya Vision 32B","description":"Open multilingual vision model for OCR, visual reasoning, and image question answering","canonical_model_id":"cohere/c4ai-aya-vision-32b","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":16000,"output":4000},"release_date":"2025-03-04","last_updated":"2025-05-14","open_weights":true},
  "command-a-03-2025": {"name":"Command A","description":"Cohere command model for multilingual enterprise agents, tools, and chat","family":"command-a","canonical_model_id":"cohere/command-a-03-2025","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":8000},"cost":{"input":2.5,"output":10},"knowledge":"2024-06-01","release_date":"2025-03-13","last_updated":"2025-03-13","open_weights":true},
  "command-a-plus-05-2026": {"name":"Command A Plus","description":"Cohere's stronger command model for multilingual agents and enterprise workflows","family":"command-a","canonical_model_id":"cohere/command-a-plus-05-2026","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]},{"type":"budget_tokens","min":1},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":64000},"cost":{"input":2.5,"output":10},"knowledge":"2025-04-01","release_date":"2026-05-20","last_updated":"2026-06-09","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "command-a-reasoning-08-2025": {"name":"Command A Reasoning","description":"Cohere reasoning model for multilingual enterprise agents, tools, and complex workflows","family":"command-a","canonical_model_id":"cohere/command-a-reasoning-08-2025","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]},{"type":"budget_tokens","min":1},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":32000},"cost":{"input":2.5,"output":10},"knowledge":"2024-06-01","release_date":"2025-08-21","last_updated":"2025-08-21","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "command-a-translate-08-2025": {"name":"Command A Translate","description":"Translation model for multilingual conversion, localization, and cross-language workflows","family":"command-a","canonical_model_id":"cohere/command-a-translate-08-2025","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8000,"output":8000},"cost":{"input":2.5,"output":10},"knowledge":"2024-06-01","release_date":"2025-08-28","last_updated":"2025-08-28","open_weights":true},
  "command-a-vision-07-2025": {"name":"Command A Vision","description":"Cohere vision model for multilingual document analysis, OCR, and image understanding","family":"command-a","canonical_model_id":"cohere/command-a-vision-07-2025","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":8000},"cost":{"input":2.5,"output":10},"knowledge":"2024-06-01","release_date":"2025-07-31","last_updated":"2025-07-31","open_weights":true},
  "command-r-08-2024": {"name":"Command R","description":"Cohere retrieval model for long-context chat and enterprise RAG workflows","family":"command-r","canonical_model_id":"cohere/command-r-08-2024","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":0.15,"output":0.6},"knowledge":"2024-06-01","release_date":"2024-08-30","last_updated":"2024-08-30","open_weights":true},
  "command-r-plus-08-2024": {"name":"Command R+","description":"Cohere's RAG workhorse for long-context enterprise search and tool use","family":"command-r","canonical_model_id":"cohere/command-r-plus-08-2024","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":2.5,"output":10},"knowledge":"2024-06-01","release_date":"2024-08-30","last_updated":"2024-08-30","open_weights":true},
  "command-r7b-12-2024": {"name":"Command R7B","description":"Cohere retrieval model for long-context chat and enterprise RAG workflows","family":"command-r","canonical_model_id":"cohere/command-r7b-12-2024","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":0.0375,"output":0.15},"knowledge":"2024-06-01","release_date":"2024-12-13","last_updated":"2024-12-13","open_weights":true},
  "command-r7b-arabic-02-2025": {"name":"Command R7B Arabic","description":"Open Command R model optimized for Arabic enterprise chat, RAG, and cultural knowledge","family":"command-r","canonical_model_id":"cohere/command-r7b-arabic-02-2025","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":0.0375,"output":0.15},"knowledge":"2024-06-01","release_date":"2025-02-27","last_updated":"2025-02-27","open_weights":true},
  "embed-english-light-v3.0": {"name":"Embed English Light v3","modalities":{"input":["text","image"]},"limit":{"context":512},"rig":{"pinned":["name","modalities","limit.context"]}},
  "embed-english-v3.0": {"name":"Embed English v3","modalities":{"input":["text","image"]},"limit":{"context":512},"rig":{"pinned":["name","modalities","limit.context"]}},
  "embed-multilingual-light-v3.0": {"name":"Embed Multilingual Light v3","modalities":{"input":["text","image"]},"limit":{"context":512},"rig":{"pinned":["name","modalities","limit.context"]}},
  "embed-multilingual-v3.0": {"name":"Embed Multilingual v3","modalities":{"input":["text","image"]},"limit":{"context":512},"rig":{"pinned":["name","modalities","limit.context"]}},
  "embed-v4.0": {"name":"Embed v4","modalities":{"input":["text","image"]},"limit":{"context":128000},"rig":{"pinned":["name","modalities","limit.context"]}},
  "north-mini-code-1-0": {"name":"North Mini Code","description":"Cohere coding model for practical software engineering and agentic edits","family":"north","canonical_model_id":"cohere/north-mini-code-1-0","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0},"knowledge":"2025-09-23","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://api.cohere.ai/compatibility/v1"}},
  "north-small-translate-09-2026": {"name":"North Small Translate","description":"Open-weight translation model for more than 50 languages and locale variants","family":"north","canonical_model_id":"cohere/north-small-translate-1-0","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":16000},"cost":{"input":0,"output":0},"release_date":"2026-09-09","last_updated":"2026-09-09","open_weights":true},
  "tiny-aya-earth": {"name":"Tiny Aya Earth","description":"Compact multilingual model specialized for West Asian and African languages","canonical_model_id":"cohere/tiny-aya-earth","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8000,"output":8000},"release_date":"2026-02-17","last_updated":"2026-02-17","open_weights":true},
  "tiny-aya-fire": {"name":"Tiny Aya Fire","description":"Compact multilingual model specialized for South Asian languages","canonical_model_id":"cohere/tiny-aya-fire","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8000,"output":8000},"release_date":"2026-02-17","last_updated":"2026-02-17","open_weights":true},
  "tiny-aya-global": {"name":"Tiny Aya Global","description":"Compact multilingual model balanced across 70 languages and regions","canonical_model_id":"cohere/tiny-aya-global","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8000,"output":8000},"release_date":"2026-02-17","last_updated":"2026-02-17","open_weights":true},
  "tiny-aya-water": {"name":"Tiny Aya Water","description":"Compact multilingual model specialized for European and Asia-Pacific languages","canonical_model_id":"cohere/tiny-aya-water","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8000,"output":8000},"release_date":"2026-02-17","last_updated":"2026-02-17","open_weights":true}
}},
"copilot": {"models": {
  "claude-3.5-sonnet": {"name":"Claude 3.5 Sonnet","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "claude-fable-5": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-fable-5.1": {"name":"Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-haiku-4.5": {"name":"Claude Haiku 4.5 (latest)","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024,"max":32000}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"input":136000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-haiku-5.5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-opus-4.6": {"name":"Claude Opus 4.6","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "claude-opus-4.7": {"name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"input":168000,"output":32000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"},"experimental":{"modes":{"fast":{"cost":{"input":30,"output":150,"cache_read":3,"cache_write":37.5},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}}},
  "claude-opus-4.8": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"input":168000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"},"experimental":{"modes":{"fast":{"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"provider":{"body":{"speed":"fast"},"headers":{"anthropic-beta":"fast-mode-2026-02-01"}}}}}},
  "claude-opus-5": {"name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"input":936000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-opus-5.5": {"name":"Claude Opus 5.5","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-sonnet-4": {"name":"Claude Sonnet 4","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "claude-sonnet-4.6": {"name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024,"max":32000}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"input":168000,"output":32000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-sonnet-5": {"name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "claude-sonnet-5.5": {"name":"Claude Sonnet 5.5","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://api.githubcopilot.com/v1"}},
  "gemini-2.0-flash-001": {"name":"Gemini 2.0 Flash","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gemini-3-flash-preview": {"name":"Gemini 3 Flash Preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "gemini-3.1-pro-preview": {"name":"Gemini 3.1 Pro Preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "gemini-3.5-flash": {"name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.5-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]},{"type":"budget_tokens","min":256,"max":24000}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":200000,"input":128000,"output":64000},"cost":{"input":1.5,"output":9,"cache_read":0.15,"input_audio":1.5},"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","open_weights":false,"status":"deprecated"},
  "gemini-3.6-flash": {"name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]},{"type":"budget_tokens","min":256,"max":32000}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"input":936000,"output":64000},"cost":{"input":0.75,"output":3.75,"cache_read":0.075},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"status":"deprecated"},
  "gemini-3.7-flash": {"name":"Gemini 3.7 Flash","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","canonical_model_id":"google/gemini-3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"input":936000,"output":64000},"cost":{"input":0.75,"output":3.75,"cache_read":0.075},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false,"status":"deprecated"},
  "gemini-3.8-flash": {"name":"Gemini 3.8 Flash","description":"Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows","family":"gemini-flash","canonical_model_id":"google/gemini-3.8-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"input":936000,"output":64000},"cost":{"input":0.75,"output":3.75,"cache_read":0.075},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false},
  "gemini-4-argon": {"name":"Gemini 4 Argon","description":"Google's flagship reasoning model for complex software engineering, deep math and science, and long-horizon autonomous agents","family":"gemini","canonical_model_id":"google/gemini-4-argon","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":10,"cache_read":0.1},"knowledge":"2026-03","release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "gpt-4": {"name":"GPT-4","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gpt-4.1": {"name":"GPT-4.1","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gpt-4.1-mini": {"name":"GPT-4.1 mini","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gpt-4.1-nano": {"name":"GPT-4.1 nano","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gpt-4o": {"name":"GPT-4o","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gpt-4o-mini": {"name":"GPT-4o mini","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gpt-5-mini": {"name":"GPT-5 Mini","description":"Small GPT-5 for responsive agents, coding help, and everyday automation","family":"gpt-mini","canonical_model_id":"openai/gpt-5-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":264000,"input":128000,"output":64000},"cost":{"input":0.25,"output":2,"cache_read":0.025},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"status":"deprecated","rig":{"sampling":"never"}},
  "gpt-5.1-codex": {"name":"GPT-5.1-Codex","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "gpt-5.3-codex": {"name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.3-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","canonical_model_id":"openai/gpt-5.4","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/openai"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-mini": {"name":"GPT-5.4 mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","canonical_model_id":"openai/gpt-5.4-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/openai"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-nano": {"name":"GPT-5.4 nano","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","canonical_model_id":"openai/gpt-5.4-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.5": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","canonical_model_id":"openai/gpt-5.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/openai"}},
  "gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-sol": {"name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-terra": {"name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-luna": {"name":"GPT-6 Luna","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-sol": {"name":"GPT-6 Sol","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-sol": {"name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-terra": {"name":"GPT-6.1 Terra","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false,"provider":{"npm":"@ai-sdk/openai"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "grok-4.5": {"name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"input":372000,"output":128000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"release_date":"2026-07-08","last_updated":"2026-07-08","open_weights":false,"provider":{"npm":"@ai-sdk/openai"}},
  "grok-4.6": {"name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"input":372000,"output":128000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":false,"provider":{"npm":"@ai-sdk/openai"}},
  "grok-4.7": {"name":"Grok 4.7","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"input":372000,"output":128000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false,"provider":{"npm":"@ai-sdk/openai"}},
  "kimi-k2.7-code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"input":224000,"output":32000},"cost":{"input":0.95,"output":4,"cache_read":0.19},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true,"status":"deprecated"},
  "kimi-k3": {"name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "mai-code-1.1-flash": {"name":"MAI-Code-1.1-Flash","description":"Microsoft coding model with native vision support, optimized for fast and efficient software development","family":"mai","canonical_model_id":"microsoft/mai-code-1.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":256000,"input":128000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02},"release_date":"2026-08-11","last_updated":"2026-08-11","open_weights":false,"provider":{"npm":"@ai-sdk/openai"}},
  "o3-mini": {"name":"o3-mini","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "text-embedding-3-large": {"name":"text-embedding-3-large","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "text-embedding-3-small": {"name":"text-embedding-3-small","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "text-embedding-ada-002": {"name":"text-embedding-ada-002","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}}
}},
"deepseek": {"models": {
  "deepseek-flash": {"name":"DeepSeek V4.1 Flash 1017","description":"October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.15,"output":0.6,"reasoning":0.6,"cache_read":0.003},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true,"rig":{"cache":["short"],"reasoning_field":"reasoning_content"}},
  "deepseek-v4-flash": {"name":"DeepSeek V4 Flash","description":"October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.15,"output":0.6,"reasoning":0.6,"cache_read":0.003},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true,"status":"deprecated","rig":{"cache":["short"],"reasoning_field":"reasoning_content"}},
  "deepseek-v4-flash-vision-exp": {"name":"DeepSeek V4 Flash Vision Exp","description":"October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.15,"output":0.6,"reasoning":0.6,"cache_read":0.003},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true,"status":"deprecated","rig":{"cache":["short"],"reasoning_field":"reasoning_content"}},
  "deepseek-v4-pro": {"name":"DeepSeek V4 Pro","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro-0813","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.66,"output":1.98,"reasoning":1.98,"cache_read":0.022},"release_date":"2026-08-13","last_updated":"2026-08-22","open_weights":true,"rig":{"cache":["short"],"reasoning_field":"reasoning_content"}}
}},
"doubleword": {"models": {
  "Qwen/Qwen3-Embedding-8B": {"name":"Qwen3 Embedding 8B","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "Qwen/Qwen3-VL-235B-A22B-Instruct-FP8": {"name":"Qwen3 VL 235B","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "Qwen/Qwen3-VL-30B-A3B-Instruct-FP8": {"name":"Qwen3 VL 30B","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "Qwen/Qwen3.5-397B-A17B-FP8": {"name":"Qwen3.5 397B A17B","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "Qwen/Qwen3.5-4B": {"name":"Qwen3.5 4B","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "Qwen/Qwen3.5-9B": {"name":"Qwen3.5 9B","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "Qwen/Qwen3.6-35B-A3B-FP8": {"name":"Qwen3.6 35B A3B","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "deepseek-ai/DeepSeek-V4-Flash": {"name":"DeepSeek V4 Flash","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "deepseek-ai/DeepSeek-V4-Pro": {"name":"DeepSeek V4 Pro","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "moonshotai/Kimi-K2.6": {"name":"Kimi K2.6","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "openai/gpt-oss-120b": {"name":"gpt-oss-120b","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "openai/gpt-oss-20b": {"name":"gpt-oss-20b","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "zai-org/GLM-5.2-FP8": {"name":"GLM 5.2","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}}
}},
"gcp.gemini": {"models": {
  "deep-research-max-preview-04-2026": {"name":"Deep Research Max Preview (Apr-21-2026)","description":"Maximum-comprehensiveness agentic researcher for multi-step investigation, synthesis, and cited reports","family":"gemini-pro","canonical_model_id":"google/deep-research-max-preview-04-2026","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false},
  "deep-research-preview-04-2026": {"name":"Deep Research Preview (Apr-21-2026)","description":"Agentic model for autonomous multi-step research, synthesis, and cited reports","family":"gemini-pro","canonical_model_id":"google/deep-research-preview-04-2026","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false},
  "gemini-2.0-flash": {"name":"Gemini 2.0 Flash","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image","audio","video","pdf"]},"limit":{"context":1048576,"output":8192},"cost":{"input":0.1,"output":0.4,"cache_read":0.025},"status":"deprecated","rig":{"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","status"]}},
  "gemini-2.0-flash-lite": {"name":"Gemini 2.0 Flash-Lite","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image","audio","video","pdf"]},"limit":{"context":1048576,"output":8192},"cost":{"input":0.075,"output":0.3},"status":"deprecated","rig":{"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","status"]}},
  "gemini-2.5-computer-use-preview-10-2025": {"name":"Gemini 2.5 Computer Use Preview 10-2025","description":"Specialized Gemini 2.5 model for browser-control agents that automate UI tasks","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-computer-use-preview-10-2025","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":64000},"cost":{"input":1.25,"output":10,"tiers":[{"input":2.5,"output":15,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15}},"knowledge":"2025-01","release_date":"2025-10-07","last_updated":"2025-10-07","open_weights":false},
  "gemini-2.5-flash": {"name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","canonical_model_id":"google/gemini-2.5-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"input_audio":1},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false},
  "gemini-2.5-flash-image": {"name":"Nano Banana","description":"Nano Banana image model for fast generation, edits, and character-consistent assets","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-2.5-flash-image","reasoning":false,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":32768,"output":32768},"cost":{"input":0.3,"output":30,"cache_read":0.075},"knowledge":"2024-06","release_date":"2025-08-26","last_updated":"2025-08-26","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning","reasoning_options"]}},
  "gemini-2.5-flash-lite": {"name":"Gemini 2.5 Flash-Lite","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":512,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01,"input_audio":0.3},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false},
  "gemini-2.5-flash-preview-04-17": {"name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","canonical_model_id":"google/gemini-2.5-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"input_audio":1},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.input_audio","knowledge","release_date","last_updated","open_weights","status"]}},
  "gemini-2.5-flash-preview-tts": {"name":"Gemini 2.5 Flash Preview TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gemini-flash","type":"speech","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":8192,"output":16384},"cost":{"input":0.5,"output":10},"knowledge":"2025-01","release_date":"2025-05-01","last_updated":"2025-05-01","open_weights":false},
  "gemini-2.5-pro": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false},
  "gemini-2.5-pro-exp-03-25": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","status"]}},
  "gemini-2.5-pro-preview-03-25": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","status"]}},
  "gemini-2.5-pro-preview-05-06": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","status"]}},
  "gemini-2.5-pro-preview-06-05": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","cost.tiers","cost.context_over_200k","knowledge","release_date","last_updated","open_weights","status"]}},
  "gemini-2.5-pro-preview-tts": {"name":"Gemini 2.5 Pro Preview TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gemini-flash","type":"speech","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":8192,"output":16384},"cost":{"input":1,"output":20},"knowledge":"2025-01","release_date":"2025-05-01","last_updated":"2025-05-01","open_weights":false},
  "gemini-3-flash-preview": {"name":"Gemini 3 Flash Preview","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"input_audio":1},"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","open_weights":false,"rig":{"reasoning_default":"high"}},
  "gemini-3-pro-image": {"name":"Nano Banana Pro","description":"Nano Banana Pro for higher-fidelity image generation and design-heavy edits","family":"gemini-pro","type":"image","canonical_model_id":"google/gemini-3-pro-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high"]}],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":65536,"output":32768},"cost":{"input":2,"output":120},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "gemini-3-pro-preview": {"name":"Gemini 3 Pro Preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image","audio","video","pdf"]},"limit":{"context":1048576,"output":65536},"status":"deprecated","rig":{"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","status"],"reasoning_default":"high"}},
  "gemini-3.1-flash-image": {"name":"Nano Banana 2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-3.1-flash-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.5,"output":60},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "gemini-3.1-flash-lite": {"name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5},"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","open_weights":false},
  "gemini-3.1-flash-lite-image": {"name":"Nano Banana 2 Lite","description":"Fastest, most cost-efficient Gemini image model for high-volume 1K generation and editing","family":"gemini-flash-lite","type":"image","canonical_model_id":"google/gemini-3.1-flash-lite-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":65536,"output":4096},"cost":{"input":0.25,"output":30},"knowledge":"2025-01","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false},
  "gemini-3.1-flash-lite-preview": {"name":"Gemini 3.1 Flash Lite Preview","description":"Legacy model retained for compatibility with older integrations","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5},"knowledge":"2025-01","release_date":"2026-03-03","last_updated":"2026-03-03","open_weights":false,"status":"deprecated"},
  "gemini-3.1-flash-live-preview": {"name":"Gemini 3.1 Flash Live Preview","description":"High-quality, low-latency Live API model for real-time dialogue and voice-first AI applications","family":"gemini-flash","type":"realtime","canonical_model_id":"google/gemini-3.1-flash-live-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio"],"output":["text","audio"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.75,"output":4.5,"input_audio":3,"output_audio":12},"knowledge":"2025-01","release_date":"2026-03-26","last_updated":"2026-03-26","open_weights":false},
  "gemini-3.1-flash-tts-preview": {"name":"Gemini 3.1 Flash TTS Preview","description":"Low-latency speech generation with steerable prompts and expressive audio tags","family":"gemini-flash","type":"speech","canonical_model_id":"google/gemini-3.1-flash-tts-preview","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":8192,"output":16384},"cost":{"input":1,"output":20},"knowledge":"2025-01","release_date":"2026-04-15","last_updated":"2026-04-15","open_weights":false},
  "gemini-3.1-pro-preview": {"name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false,"rig":{"reasoning_default":"high"}},
  "gemini-3.1-pro-preview-customtools": {"name":"Gemini 3.1 Pro Preview Custom Tools","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false},
  "gemini-3.5-flash": {"name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"input_audio":1.5},"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.5-flash-lite": {"name":"Gemini 3.5 Flash Lite","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.5-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"rig":{"reasoning_default":"minimal"}},
  "gemini-3.5-live-translate-preview": {"name":"Gemini 3.5 Live Translate Preview","description":"Low-latency audio-to-audio model for real-time speech translation across 70+ languages","family":"gemini-pro","type":"realtime","canonical_model_id":"google/gemini-3.5-live-translate-preview","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["audio"],"output":["audio","text"]},"limit":{"context":16384,"output":32768},"cost":{"input":3.5,"output":21,"input_audio":3.5,"output_audio":21},"knowledge":"2025-01","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false},
  "gemini-3.6-flash": {"name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.7-flash": {"name":"Gemini 3.7 Flash","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","canonical_model_id":"google/gemini-3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.8-flash": {"name":"Gemini 3.8 Flash","description":"Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows","family":"gemini-flash","canonical_model_id":"google/gemini-3.8-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.8-flash-lite-tts": {"name":"Gemini 3.8 Flash-Lite TTS","reasoning":false,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":8192,"output":16384},"cost":{"input":0.5,"output":6},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output"]}},
  "gemini-3.8-flash-tts": {"name":"Gemini 3.8 Flash TTS","reasoning":false,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":8192,"output":16384},"cost":{"input":0.5,"output":9},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output"]}},
  "gemini-4-argon": {"name":"Gemini 4 Argon","description":"Google's flagship reasoning model for complex software engineering, deep math and science, and long-horizon autonomous agents","family":"gemini","canonical_model_id":"google/gemini-4-argon","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":32768},"cost":{"input":2,"output":10,"cache_read":0.1,"input_audio":2},"knowledge":"2026-03","release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "gemini-embedding-001": {"name":"Gemini Embedding 001","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"gemini","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":2048,"output":1},"cost":{"input":0.15,"output":0},"knowledge":"2025-05","release_date":"2025-05-20","last_updated":"2025-05-20","open_weights":false},
  "gemini-embedding-2": {"name":"Gemini Embedding 2","description":"Multimodal embedding model mapping text, images, video, audio, and PDFs into a unified embedding space","family":"gemini","type":"embedding","canonical_model_id":"google/gemini-embedding-2","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":8192,"output":1},"cost":{"input":0.2,"output":0,"input_audio":6.5},"knowledge":"2025-11","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":false},
  "gemini-flash-latest": {"name":"Gemini Flash Latest","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false},
  "gemini-flash-lite-latest": {"name":"Gemini Flash-Lite Latest","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false},
  "gemini-nano-banana-2.1": {"name":"Nano Banana 2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","video","pdf"]},"limit":{"context":131072,"output":32768},"cost":{"input":1.5,"output":30},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output"],"reasoning_default":"medium"}},
  "gemini-omni-flash-preview": {"name":"Gemini Omni Flash Preview","description":"Video generation and editing model for fast, conversational text- and image-to-video workflows","family":"gemini","type":"video","canonical_model_id":"google/gemini-omni-flash-preview","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["video"]},"limit":{"context":131072,"output":65536},"cost":{"input":1.5,"output":17.5},"release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false},
  "gemini-pro-latest": {"name":"Gemini Pro Latest","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","canonical_model_id":"google/gemini-3.1-pro-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false},
  "gemma-4-26b-a4b-it": {"name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "gemma-4-31b-it": {"name":"Gemma 4 31B IT","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "lyria-3-clip-preview": {"name":"Lyria 3 Clip Preview","description":"Music generation model for short 30-second clips, loops, and previews from text or image prompts","family":"lyria","canonical_model_id":"google/lyria-3-clip-preview","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","audio"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0,"output":0},"release_date":"2026-03-25","last_updated":"2026-03-25","open_weights":false},
  "lyria-3-pro-preview": {"name":"Lyria 3 Pro Preview","description":"Music generation model for full-length songs from text or images with vocals and structure","family":"lyria","canonical_model_id":"google/lyria-3-pro-preview","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","audio"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0,"output":0},"release_date":"2026-03-25","last_updated":"2026-03-25","open_weights":false},
  "text-embedding-004": {"name":"Text Embedding 004","modalities":{"input":["text"]},"limit":{"context":2048},"status":"deprecated","rig":{"pinned":["name","modalities","limit.context","status"]}}
}},
"gemini-grpc": {"models": {
  "deep-research-max-preview-04-2026": {"name":"Deep Research Max Preview (Apr-21-2026)","description":"Maximum-comprehensiveness agentic researcher for multi-step investigation, synthesis, and cited reports","family":"gemini-pro","canonical_model_id":"google/deep-research-max-preview-04-2026","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false},
  "deep-research-preview-04-2026": {"name":"Deep Research Preview (Apr-21-2026)","description":"Agentic model for autonomous multi-step research, synthesis, and cited reports","family":"gemini-pro","canonical_model_id":"google/deep-research-preview-04-2026","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false},
  "gemini-2.0-flash": {"name":"Gemini 2.0 Flash","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image","audio","video","pdf"]},"limit":{"context":1048576,"output":8192},"cost":{"input":0.1,"output":0.4,"cache_read":0.025},"status":"deprecated"},
  "gemini-2.0-flash-lite": {"name":"Gemini 2.0 Flash-Lite","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image","audio","video","pdf"]},"limit":{"context":1048576,"output":8192},"cost":{"input":0.075,"output":0.3},"status":"deprecated"},
  "gemini-2.5-computer-use-preview-10-2025": {"name":"Gemini 2.5 Computer Use Preview 10-2025","description":"Specialized Gemini 2.5 model for browser-control agents that automate UI tasks","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-computer-use-preview-10-2025","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":64000},"cost":{"input":1.25,"output":10,"tiers":[{"input":2.5,"output":15,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15}},"knowledge":"2025-01","release_date":"2025-10-07","last_updated":"2025-10-07","open_weights":false},
  "gemini-2.5-flash": {"name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","canonical_model_id":"google/gemini-2.5-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"input_audio":1},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false},
  "gemini-2.5-flash-image": {"name":"Nano Banana","description":"Nano Banana image model for fast generation, edits, and character-consistent assets","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-2.5-flash-image","reasoning":false,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":32768,"output":32768},"cost":{"input":0.3,"output":30,"cache_read":0.075},"knowledge":"2024-06","release_date":"2025-08-26","last_updated":"2025-08-26","open_weights":false,"status":"deprecated"},
  "gemini-2.5-flash-lite": {"name":"Gemini 2.5 Flash-Lite","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":512,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01,"input_audio":0.3},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false},
  "gemini-2.5-flash-preview-04-17": {"name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","canonical_model_id":"google/gemini-2.5-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"input_audio":1},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-flash-preview-tts": {"name":"Gemini 2.5 Flash Preview TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gemini-flash","type":"speech","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":8192,"output":16384},"cost":{"input":0.5,"output":10},"knowledge":"2025-01","release_date":"2025-05-01","last_updated":"2025-05-01","open_weights":false},
  "gemini-2.5-pro": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false},
  "gemini-2.5-pro-exp-03-25": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-pro-preview-03-25": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-pro-preview-05-06": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-pro-preview-06-05": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-pro-preview-tts": {"name":"Gemini 2.5 Pro Preview TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gemini-flash","type":"speech","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":8192,"output":16384},"cost":{"input":1,"output":20},"knowledge":"2025-01","release_date":"2025-05-01","last_updated":"2025-05-01","open_weights":false},
  "gemini-3-flash-preview": {"name":"Gemini 3 Flash Preview","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"input_audio":1},"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","open_weights":false,"rig":{"reasoning_default":"high"}},
  "gemini-3-pro-image": {"name":"Nano Banana Pro","description":"Nano Banana Pro for higher-fidelity image generation and design-heavy edits","family":"gemini-pro","type":"image","canonical_model_id":"google/gemini-3-pro-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high"]}],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":65536,"output":32768},"cost":{"input":2,"output":120},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "gemini-3-pro-preview": {"name":"Gemini 3 Pro Preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image","audio","video","pdf"]},"limit":{"context":1048576,"output":65536},"status":"deprecated","rig":{"reasoning_default":"high"}},
  "gemini-3.1-flash-image": {"name":"Nano Banana 2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-3.1-flash-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.5,"output":60},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "gemini-3.1-flash-lite": {"name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5},"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","open_weights":false},
  "gemini-3.1-flash-lite-image": {"name":"Nano Banana 2 Lite","description":"Fastest, most cost-efficient Gemini image model for high-volume 1K generation and editing","family":"gemini-flash-lite","type":"image","canonical_model_id":"google/gemini-3.1-flash-lite-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":65536,"output":4096},"cost":{"input":0.25,"output":30},"knowledge":"2025-01","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false},
  "gemini-3.1-flash-lite-preview": {"name":"Gemini 3.1 Flash Lite Preview","description":"Legacy model retained for compatibility with older integrations","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5},"knowledge":"2025-01","release_date":"2026-03-03","last_updated":"2026-03-03","open_weights":false,"status":"deprecated"},
  "gemini-3.1-flash-live-preview": {"name":"Gemini 3.1 Flash Live Preview","description":"High-quality, low-latency Live API model for real-time dialogue and voice-first AI applications","family":"gemini-flash","type":"realtime","canonical_model_id":"google/gemini-3.1-flash-live-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio"],"output":["text","audio"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.75,"output":4.5,"input_audio":3,"output_audio":12},"knowledge":"2025-01","release_date":"2026-03-26","last_updated":"2026-03-26","open_weights":false},
  "gemini-3.1-flash-tts-preview": {"name":"Gemini 3.1 Flash TTS Preview","description":"Low-latency speech generation with steerable prompts and expressive audio tags","family":"gemini-flash","type":"speech","canonical_model_id":"google/gemini-3.1-flash-tts-preview","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":8192,"output":16384},"cost":{"input":1,"output":20},"knowledge":"2025-01","release_date":"2026-04-15","last_updated":"2026-04-15","open_weights":false},
  "gemini-3.1-pro-preview": {"name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false,"rig":{"reasoning_default":"high"}},
  "gemini-3.1-pro-preview-customtools": {"name":"Gemini 3.1 Pro Preview Custom Tools","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false},
  "gemini-3.5-flash": {"name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"input_audio":1.5},"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.5-flash-lite": {"name":"Gemini 3.5 Flash Lite","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.5-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"rig":{"reasoning_default":"minimal"}},
  "gemini-3.5-live-translate-preview": {"name":"Gemini 3.5 Live Translate Preview","description":"Low-latency audio-to-audio model for real-time speech translation across 70+ languages","family":"gemini-pro","type":"realtime","canonical_model_id":"google/gemini-3.5-live-translate-preview","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["audio"],"output":["audio","text"]},"limit":{"context":16384,"output":32768},"cost":{"input":3.5,"output":21,"input_audio":3.5,"output_audio":21},"knowledge":"2025-01","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false},
  "gemini-3.6-flash": {"name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.7-flash": {"name":"Gemini 3.7 Flash","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","canonical_model_id":"google/gemini-3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.8-flash": {"name":"Gemini 3.8 Flash","description":"Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows","family":"gemini-flash","canonical_model_id":"google/gemini-3.8-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"rig":{"reasoning_default":"medium"}},
  "gemini-3.8-flash-lite-tts": {"name":"Gemini 3.8 Flash-Lite TTS","reasoning":false,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":8192,"output":16384},"cost":{"input":0.5,"output":6}},
  "gemini-3.8-flash-tts": {"name":"Gemini 3.8 Flash TTS","reasoning":false,"reasoning_options":[],"tool_call":false,"structured_output":false,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":8192,"output":16384},"cost":{"input":0.5,"output":9}},
  "gemini-4-argon": {"name":"Gemini 4 Argon","description":"Google's flagship reasoning model for complex software engineering, deep math and science, and long-horizon autonomous agents","family":"gemini","canonical_model_id":"google/gemini-4-argon","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":32768},"cost":{"input":2,"output":10,"cache_read":0.1,"input_audio":2},"knowledge":"2026-03","release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "gemini-embedding-001": {"name":"Gemini Embedding 001","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"gemini","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":2048,"output":1},"cost":{"input":0.15,"output":0},"knowledge":"2025-05","release_date":"2025-05-20","last_updated":"2025-05-20","open_weights":false},
  "gemini-embedding-2": {"name":"Gemini Embedding 2","description":"Multimodal embedding model mapping text, images, video, audio, and PDFs into a unified embedding space","family":"gemini","type":"embedding","canonical_model_id":"google/gemini-embedding-2","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":8192,"output":1},"cost":{"input":0.2,"output":0,"input_audio":6.5},"knowledge":"2025-11","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":false},
  "gemini-flash-latest": {"name":"Gemini Flash Latest","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false},
  "gemini-flash-lite-latest": {"name":"Gemini Flash-Lite Latest","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false},
  "gemini-nano-banana-2.1": {"name":"Nano Banana 2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":true,"modalities":{"input":["text","image","video","pdf"]},"limit":{"context":131072,"output":32768},"cost":{"input":1.5,"output":30},"rig":{"reasoning_default":"medium"}},
  "gemini-omni-flash-preview": {"name":"Gemini Omni Flash Preview","description":"Video generation and editing model for fast, conversational text- and image-to-video workflows","family":"gemini","type":"video","canonical_model_id":"google/gemini-omni-flash-preview","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["video"]},"limit":{"context":131072,"output":65536},"cost":{"input":1.5,"output":17.5},"release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false},
  "gemini-pro-latest": {"name":"Gemini Pro Latest","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","canonical_model_id":"google/gemini-3.1-pro-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false},
  "gemma-4-26b-a4b-it": {"name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true},
  "gemma-4-31b-it": {"name":"Gemma 4 31B IT","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true},
  "lyria-3-clip-preview": {"name":"Lyria 3 Clip Preview","description":"Music generation model for short 30-second clips, loops, and previews from text or image prompts","family":"lyria","canonical_model_id":"google/lyria-3-clip-preview","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","audio"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0,"output":0},"release_date":"2026-03-25","last_updated":"2026-03-25","open_weights":false},
  "lyria-3-pro-preview": {"name":"Lyria 3 Pro Preview","description":"Music generation model for full-length songs from text or images with vocals and structure","family":"lyria","canonical_model_id":"google/lyria-3-pro-preview","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","audio"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0,"output":0},"release_date":"2026-03-25","last_updated":"2026-03-25","open_weights":false},
  "text-embedding-004": {"name":"Text Embedding 004","modalities":{"input":["text"]},"limit":{"context":2048},"status":"deprecated"}
}},
"groq": {"models": {
  "allam-2-7b": {"name":"ALLaM-2-7b","description":"ALLaM-2-7b instruction tuned model by SDAIA","canonical_model_id":"sdaia/allam-2-7b","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":4096,"output":4096},"cost":{"input":0,"output":0},"release_date":"2025-01-23","last_updated":"2025-01-23","open_weights":true},
  "canopylabs/orpheus-arabic-saudi": {"name":"Canopy Labs Orpheus Arabic Saudi","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"canopylabs","type":"speech","canonical_model_id":"canopylabs/orpheus-arabic-saudi","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":4000,"output":50000},"release_date":"2025-12-16","last_updated":"2025-12-16","open_weights":false,"status":"beta","provider":{"npm":"@ai-sdk/openai","api":"https://api.groq.com/openai/v1"}},
  "canopylabs/orpheus-v1-english": {"name":"Canopy Labs Orpheus V1 English","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"canopylabs","type":"speech","canonical_model_id":"canopylabs/orpheus-v1-english","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":4000,"output":50000},"release_date":"2025-12-19","last_updated":"2025-12-19","open_weights":false,"status":"beta","provider":{"npm":"@ai-sdk/openai","api":"https://api.groq.com/openai/v1"}},
  "llama-3.1-8b-instant": {"name":"Llama 3.1 8B","description":"Compact Llama instruction model for fast chat and local deployment","family":"llama","canonical_model_id":"meta/llama-3.1-8b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.05,"output":0.08},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true,"status":"deprecated"},
  "llama-3.3-70b-versatile": {"name":"Llama 3.3 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.3-70b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.59,"output":0.79},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true,"status":"deprecated"},
  "meta-llama/llama-prompt-guard-2-22m": {"name":"Llama Prompt Guard 2 22M","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"llama","canonical_model_id":"meta/llama-prompt-guard-2-22m","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":512,"output":512},"cost":{"input":0.03,"output":0.03},"release_date":"2025-04-29","last_updated":"2025-04-29","open_weights":true,"status":"beta"},
  "meta-llama/llama-prompt-guard-2-86m": {"name":"Prompt Guard 2 86M","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"llama","canonical_model_id":"meta/llama-prompt-guard-2-86m","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":512,"output":512},"cost":{"input":0.04,"output":0.04},"release_date":"2025-04-29","last_updated":"2025-04-29","open_weights":true,"status":"beta"},
  "minimaxai/minimax-m2.7": {"name":"MiniMax M2.7","reasoning":true,"reasoning_options":[],"tool_call":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","modalities"]}},
  "openai/gpt-oss-120b": {"name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.15,"output":0.6,"cache_read":0.075},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "openai/gpt-oss-20b": {"name":"GPT OSS 20B","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.075,"output":0.3,"cache_read":0.0375},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "openai/gpt-oss-safeguard-20b": {"name":"Safety GPT OSS 20B","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-safeguard-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.075,"output":0.3,"cache_read":0.0375},"release_date":"2025-10-29","last_updated":"2025-10-29","open_weights":true,"status":"beta"},
  "qwen/qwen3.6-27b": {"name":"Qwen3.6 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.6-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","default"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.6,"output":3,"cache_read":0.3},"release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"status":"deprecated"},
  "qwen/qwen3.8-27b": {"name":"Qwen3.8 27B","description":"Dense 27B vision-language model for coding, agent tasks, and image and video understanding","family":"qwen","canonical_model_id":"alibaba/qwen3.8-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","default","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.8,"output":4},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true,"status":"beta"},
  "whisper-large-v3": {"name":"Whisper","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"whisper","type":"transcription","canonical_model_id":"openai/whisper-large-v3","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["audio"],"output":["text"]},"limit":{"context":448,"output":448},"release_date":"2023-11-06","last_updated":"2023-11-06","open_weights":true},
  "whisper-large-v3-turbo": {"name":"Whisper Large V3 Turbo","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"whisper","type":"transcription","canonical_model_id":"openai/whisper-large-v3-turbo","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["audio"],"output":["text"]},"limit":{"context":448,"output":448},"release_date":"2024-10-01","last_updated":"2024-10-01","open_weights":true}
}},
"huggingface": {"models": {
  "Kwai-Kolors/Kolors": {"name":"Kolors","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "MiniMaxAI/MiniMax-M1-80k": {"name":"MiniMax M1 80K","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/minimax-m1-80k","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":80000},"cost":{"input":0.55,"output":2.2},"release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":true},
  "MiniMaxAI/MiniMax-M2": {"name":"MiniMax-M2","description":"Efficient open MiniMax model built for coding agents and tool-heavy workflows","family":"minimax","canonical_model_id":"minimax/MiniMax-M2","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2},"release_date":"2025-10-27","last_updated":"2025-10-27","open_weights":true},
  "MiniMaxAI/MiniMax-M2.1": {"name":"MiniMax-M2.1","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2},"knowledge":"2025-10","release_date":"2025-12-23","last_updated":"2025-12-23","open_weights":true},
  "MiniMaxAI/MiniMax-M2.5": {"name":"MiniMax-M2.5","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true},
  "MiniMaxAI/MiniMax-M2.7": {"name":"MiniMax-M2.7","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.7","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true},
  "MiniMaxAI/MiniMax-M3": {"name":"MiniMax-M3","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","canonical_model_id":"minimax/MiniMax-M3","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":524288,"output":512000},"cost":{"input":0.3,"output":1.2},"release_date":"2026-06-01","last_updated":"2026-06-01","open_weights":true},
  "MiniMaxAI/MiniMax-M3.1-Flash": {"name":"MiniMax-M3.1-Flash","description":"Frontier multimodal coding model with 1M context window and tunable thinking depth","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":512000},"cost":{"input":0.15,"output":0.6},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false},
  "PowerInfer/SmallThinker-3B-Preview": {"name":"SmallThinker 3B Preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "Qwen/QVQ-72B-Preview": {"name":"QVQ 72B Preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "Qwen/Qwen2-VL-7B-Instruct": {"name":"Qwen2 VL 7B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "Qwen/Qwen2.5-7B-Instruct": {"name":"Qwen2.5 7B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "Qwen/Qwen2.5-Coder-32B-Instruct": {"name":"Qwen2.5-Coder-32B-Instruct","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","canonical_model_id":"alibaba/qwen2.5-coder-32b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":0.06,"output":0.2},"release_date":"2024-11-12","last_updated":"2024-11-12","open_weights":true},
  "Qwen/Qwen3-235B-A22B": {"name":"Qwen3 235B-A22B","description":"Large open Qwen MoE for multilingual reasoning, coding, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":40960,"output":16384},"cost":{"input":0.2,"output":0.8},"knowledge":"2025-04","release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":true},
  "Qwen/Qwen3-235B-A22B-Instruct-2507": {"name":"Qwen3 235B-A22B Instruct 2507","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-instruct-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":0.855,"output":2.565},"release_date":"2025-07-21","last_updated":"2025-07-21","open_weights":true},
  "Qwen/Qwen3-235B-A22B-Thinking-2507": {"name":"Qwen3-235B-A22B-Thinking-2507","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-thinking-2507","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":81920},"cost":{"input":0.3,"output":3},"knowledge":"2025-04","release_date":"2025-07-25","last_updated":"2025-07-25","open_weights":true},
  "Qwen/Qwen3-30B-A3B": {"name":"Qwen3 30B A3B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-30b-a3b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":40960,"output":16384},"cost":{"input":0.12,"output":0.5},"release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":true},
  "Qwen/Qwen3-32B": {"name":"Qwen3 32B","description":"Dense open Qwen model for self-hosted chat, reasoning, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3-32b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.29,"output":0.59},"knowledge":"2025-04","release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":true},
  "Qwen/Qwen3-8B": {"name":"Qwen3 8B","description":"Dense 8B Qwen3 model with switchable thinking for chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-8b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":40960,"output":32768},"cost":{"input":0.07,"output":0.18},"release_date":"2025-04-29","last_updated":"2025-04-29","open_weights":true},
  "Qwen/Qwen3-Coder-30B-A3B-Instruct": {"name":"Qwen3-Coder 30B-A3B Instruct","description":"Smaller Qwen coder for efficient local agents and repo-level fixes","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-30b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.07,"output":0.26},"knowledge":"2025-04","release_date":"2025-07-31","last_updated":"2025-07-31","open_weights":true},
  "Qwen/Qwen3-Coder-480B-A35B-Instruct": {"name":"Qwen3-Coder-480B-A35B-Instruct","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-480b-a35b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":2,"output":2},"knowledge":"2025-04","release_date":"2025-07-22","last_updated":"2025-07-22","open_weights":true},
  "Qwen/Qwen3-Coder-Next": {"name":"Qwen3-Coder-Next","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-next","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.2,"output":1.5},"knowledge":"2025-09","release_date":"2026-02-03","last_updated":"2026-02-03","open_weights":true},
  "Qwen/Qwen3-Embedding-4B": {"name":"Qwen 3 Embedding 4B","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"qwen","type":"embedding","canonical_model_id":"alibaba/qwen3-embedding-4b","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":0},"cost":{"input":0.01,"output":0},"knowledge":"2024-12","release_date":"2025-06-05","last_updated":"2025-06-05","open_weights":true,"provider":{"api":"https://router.huggingface.co/deepinfra/v1/openai"}},
  "Qwen/Qwen3-Embedding-8B": {"name":"Qwen 3 Embedding 8B","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"qwen","type":"embedding","canonical_model_id":"alibaba/qwen3-embedding-8b","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":0},"cost":{"input":0.01,"output":0},"knowledge":"2024-12","release_date":"2025-06-05","last_updated":"2025-06-05","open_weights":true,"provider":{"api":"https://router.huggingface.co/deepinfra/v1/openai"}},
  "Qwen/Qwen3-Next-80B-A3B-Instruct": {"name":"Qwen3-Next-80B-A3B-Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-next-80b-a3b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.25,"output":1},"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","open_weights":true},
  "Qwen/Qwen3-VL-235B-A22B-Instruct": {"name":"Qwen3 VL 235B A22B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3-vl-235b-a22b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.3,"output":1.5},"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","open_weights":true},
  "Qwen/Qwen3-VL-235B-A22B-Thinking": {"name":"Qwen3 VL 235B A22B Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3-vl-235b-a22b-thinking","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.98,"output":3.95},"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","open_weights":true},
  "Qwen/Qwen3-VL-30B-A3B-Instruct": {"name":"Qwen3 VL 30B A3B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3-vl-30b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.2,"output":0.7},"release_date":"2025-10-04","last_updated":"2025-10-04","open_weights":true},
  "Qwen/Qwen3.5-122B-A10B": {"name":"Qwen3.5 122B-A10B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-122b-a10b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.4,"output":3.2},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true},
  "Qwen/Qwen3.5-27B": {"name":"Qwen3.5 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-27b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.3,"output":2.4},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true},
  "Qwen/Qwen3.5-35B-A3B": {"name":"Qwen3.5 35B-A3B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-35b-a3b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.25,"output":2},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true},
  "Qwen/Qwen3.5-397B-A17B": {"name":"Qwen3.5-397B-A17B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-397b-a17b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.6,"output":3.6},"knowledge":"2025-04","release_date":"2026-02-15","last_updated":"2026-02-15","open_weights":true},
  "Qwen/Qwen3.5-9B": {"name":"Qwen3.5 9B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.5-9b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.17,"output":0.25},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true},
  "Qwen/Qwen3.6-27B": {"name":"Qwen3.6 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.6-27b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.47,"output":3.19},"release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true},
  "Qwen/Qwen3.6-35B-A3B": {"name":"Qwen3.6 35B-A3B","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","canonical_model_id":"alibaba/qwen3.6-35b-a3b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.15,"output":0.95},"release_date":"2026-04-17","last_updated":"2026-04-17","open_weights":true},
  "Qwen/Qwen3.8-2.4T-A95B": {"name":"Qwen3.8 2.4T A95B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.8-2.4t-a95b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":2.5,"output":6.25},"release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":true},
  "Qwen/Qwen3.8-27B": {"name":"Qwen3.8 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.4,"output":3},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "XiaomiMiMo/MiMo-V2.5": {"name":"MiMo-V2.5","description":"MiMo model for long-context reasoning, perception, and agentic tasks","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.4,"output":2},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true},
  "XiaomiMiMo/MiMo-V2.5-Pro": {"name":"MiMo-V2.5-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1,"output":3},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true},
  "black-forest-labs/FLUX.1-dev": {"name":"FLUX.1 dev","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "deepseek-ai/DeepSeek-R1": {"name":"DeepSeek-R1","description":"Classic open reasoning model for transparent math, coding, and deliberate problem solving","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-r1","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":64000,"output":32768},"cost":{"input":0.7,"output":2.5},"knowledge":"2024-07","release_date":"2025-01-20","last_updated":"2025-05-29","open_weights":true},
  "deepseek-ai/DeepSeek-R1-0528": {"name":"DeepSeek-R1-0528","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-r1-0528","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":32768},"cost":{"input":3,"output":5},"knowledge":"2025-05","release_date":"2025-05-28","last_updated":"2025-05-28","open_weights":true},
  "deepseek-ai/DeepSeek-V3": {"name":"DeepSeek-V3","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":64000,"output":8192},"cost":{"input":0.4,"output":1.3},"release_date":"2024-12-26","last_updated":"2024-12-26","open_weights":true},
  "deepseek-ai/DeepSeek-V3-0324": {"name":"DeepSeek V3 0324","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3-0324","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":163840},"cost":{"input":0.27,"output":1.12},"release_date":"2025-03-24","last_updated":"2025-03-24","open_weights":true},
  "deepseek-ai/DeepSeek-V3.1": {"name":"DeepSeek-V3.1","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3.1","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":8192},"cost":{"input":0.27,"output":1},"release_date":"2025-08-21","last_updated":"2025-08-21","open_weights":true},
  "deepseek-ai/DeepSeek-V3.2": {"name":"DeepSeek-V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3.2","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":65536},"cost":{"input":0.28,"output":0.4},"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","open_weights":true},
  "deepseek-ai/DeepSeek-V4-Flash": {"name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.14,"output":0.28},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true},
  "deepseek-ai/DeepSeek-V4-Flash-0731": {"name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-0731","reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.14,"output":0.28},"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","open_weights":true},
  "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp": {"name":"DeepSeek V4 Flash Vision Exp","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-vision-exp","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.44,"output":1.32},"release_date":"2026-08-21","last_updated":"2026-09-01","open_weights":true},
  "deepseek-ai/DeepSeek-V4-Pro": {"name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.435,"output":0.87,"cache_read":0.003625},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true},
  "deepseek-ai/DeepSeek-V4-Pro-0813": {"name":"DeepSeek V4 Pro 0813","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro-0813","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":1.32,"output":3.96},"release_date":"2026-08-13","last_updated":"2026-08-22","open_weights":true},
  "deepseek-ai/DeepSeek-V4.1-Flash": {"name":"DeepSeek V4.1 Flash","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.3,"output":1.2},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true},
  "deepseek-ai/DeepSeek-V4.1-Flash-1017": {"name":"DeepSeek V4.1 Flash 1017","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.3,"output":1.2},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true},
  "google/gemma-2-2b-it": {"name":"Gemma 2 2B","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "google/gemma-3-12b-it": {"name":"Gemma 3 12B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-3-12b-it","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.05,"output":0.15},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true},
  "google/gemma-3-27b-it": {"name":"Gemma 3 27B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-3-27b-it","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.08,"output":0.16},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true},
  "google/gemma-3-4b-it": {"name":"Gemma 3 4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-3-4b-it","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.05,"output":0.1},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true},
  "google/gemma-4-26B-A4B-it": {"name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-26b-a4b-it","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.13,"output":0.4},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true},
  "google/gemma-4-31B-it": {"name":"Gemma 4 31B IT","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","canonical_model_id":"google/gemma-4-31b-it","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.14,"output":0.4},"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true},
  "meta-llama/Llama-3.1-8B-Instruct": {"name":"Llama-3.1-8B-Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.1-8b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":4096},"cost":{"input":0.06,"output":0.06},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true},
  "meta-llama/Llama-3.3-70B-Instruct": {"name":"Llama-3.3-70B-Instruct","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","canonical_model_id":"meta/llama-3.3-70b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":4096},"cost":{"input":0.59,"output":0.79},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true},
  "meta-llama/Llama-Guard-4-12B": {"name":"Llama Guard 4 12B","description":"Multimodal Llama 4-based safety classifier for moderating prompts and responses","family":"llama","canonical_model_id":"meta/llama-guard-4-12b","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":163840,"output":163840},"cost":{"input":0.18,"output":0.18},"release_date":"2025-04-29","last_updated":"2025-04-29","open_weights":true},
  "meta-llama/Meta-Llama-3.1-8B-Instruct": {"name":"Llama 3.1 8B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "moonshotai/Kimi-K2-Instruct": {"name":"Kimi-K2-Instruct","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":1,"output":3},"knowledge":"2024-10","release_date":"2025-07-11","last_updated":"2025-07-11","open_weights":true},
  "moonshotai/Kimi-K2-Instruct-0905": {"name":"Kimi-K2-Instruct-0905","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2-0905","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":1,"output":3},"knowledge":"2024-10","release_date":"2025-09-05","last_updated":"2025-09-05","open_weights":true},
  "moonshotai/Kimi-K2-Thinking": {"name":"Kimi-K2-Thinking","description":"Kimi reasoning model for long-horizon research, planning, and tool use","family":"kimi-thinking","canonical_model_id":"moonshotai/kimi-k2-thinking","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.6,"output":2.5,"cache_read":0.15},"knowledge":"2024-08","release_date":"2025-11-06","last_updated":"2025-11-06","open_weights":true},
  "moonshotai/Kimi-K2.5": {"name":"Kimi-K2.5","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.6,"output":3,"cache_read":0.1},"knowledge":"2025-01","release_date":"2026-01-27","last_updated":"2026-01-27","open_weights":true},
  "moonshotai/Kimi-K2.6": {"name":"Kimi-K2.6","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.6","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.16},"knowledge":"2025-01","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true},
  "moonshotai/Kimi-K2.7-Code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true},
  "moonshotai/Kimi-K3": {"name":"Kimi K3","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":3,"output":15},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "openai/gpt-oss-120b": {"name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.25,"output":0.69},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "openai/gpt-oss-20b": {"name":"GPT OSS 20B","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.1,"output":0.5},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "openai/whisper-large-v3": {"name":"Whisper large v3","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "openai/whisper-large-v3-turbo": {"name":"Whisper large v3 turbo","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "openai/whisper-small": {"name":"Whisper small","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "stabilityai/stable-diffusion-3-medium-diffusers": {"name":"Stable Diffusion 3 Medium","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "stepfun-ai/Step-3.5-Flash": {"name":"Step 3.5 Flash","description":"StepFun flash lane for quick multimodal reasoning and coding assistance","family":"step","canonical_model_id":"stepfun/step-3.5-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":256000},"cost":{"input":0.1,"output":0.3},"knowledge":"2025-01","release_date":"2026-01-29","last_updated":"2026-02-13","open_weights":true},
  "stepfun-ai/Step-3.7-Flash": {"name":"Step 3.7 Flash","description":"Newer StepFun flash model for faster agents, coding, and multimodal prompts","canonical_model_id":"stepfun/step-3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":256000},"cost":{"input":0.2,"output":1.15},"knowledge":"2026-03-01","release_date":"2026-05-29","last_updated":"2026-05-29","open_weights":true},
  "tencent/Hy3": {"name":"Hy3","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"input":192000,"output":128000},"cost":{"input":0.14,"output":0.58},"release_date":"2026-07-06","last_updated":"2026-07-06","open_weights":true},
  "tencent/Hy4-preview": {"name":"Hy4 preview","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy4-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":0.834,"output":2.501},"release_date":"2026-08-28","last_updated":"2026-08-28","open_weights":true},
  "thinkingmachines/Inkling": {"name":"Inkling","description":"Multimodal model for analyzing text, images, documents, and rich media","family":"ling","canonical_model_id":"thinkingmachines/inkling","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":1048576},"cost":{"input":1,"output":4.05},"release_date":"2026-07-15","last_updated":"2026-07-15","open_weights":true},
  "thinkingmachines/Inkling-Small": {"name":"Inkling Small","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","canonical_model_id":"thinkingmachines/inkling-small","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":524288,"output":1048576},"cost":{"input":0.5,"output":1.2},"release_date":"2026-07-30","last_updated":"2026-07-30","open_weights":true},
  "zai-org/AutoGLM-Phone-9B-Multilingual": {"name":"AutoGLM Phone 9B Multilingual","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","canonical_model_id":"zhipuai/autoglm-phone-9b-multilingual","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":65536,"output":65536},"cost":{"input":0.035,"output":0.138},"release_date":"2025-12-10","last_updated":"2025-12-10","open_weights":true},
  "zai-org/GLM-4.5-Air": {"name":"GLM-4.5-Air","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","canonical_model_id":"zhipuai/glm-4.5-air","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":98304},"cost":{"input":0.13,"output":0.85},"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":true},
  "zai-org/GLM-4.5V": {"name":"GLM-4.5V","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","canonical_model_id":"zhipuai/glm-4.5v","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":65536,"output":16384},"cost":{"input":0.6,"output":1.8},"knowledge":"2025-04","release_date":"2025-08-11","last_updated":"2025-08-11","open_weights":true},
  "zai-org/GLM-4.6": {"name":"GLM-4.6","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.6","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.55,"output":2.2},"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","open_weights":true},
  "zai-org/GLM-4.6V-Flash": {"name":"GLM-4.6V-Flash","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.6v-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.3,"output":0.9},"release_date":"2025-12-08","last_updated":"2025-12-08","open_weights":true},
  "zai-org/GLM-4.7": {"name":"GLM-4.7","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-4.7","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.11},"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","open_weights":true},
  "zai-org/GLM-4.7-Flash": {"name":"GLM-4.7-Flash","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.7-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":128000},"cost":{"input":0,"output":0},"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","open_weights":true},
  "zai-org/GLM-5": {"name":"GLM-5","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true},
  "zai-org/GLM-5.1": {"name":"GLM-5.1","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-5.1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2},"release_date":"2026-04-07","last_updated":"2026-04-07","open_weights":true},
  "zai-org/GLM-5.2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":1.4,"output":4.4},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true},
  "zai-org/GLM-5.3": {"name":"GLM-5.3","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "zai-org/GLM-5.3-Flash": {"name":"GLM-5.3-Flash","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.15,"output":0.5},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true}
}},
"hyperbolic": {"models": {
  "NousResearch/Hermes-3-Llama-3.1-70b": {"name":"Hermes 3 70B","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "Qwen/QwQ-32B-Preview": {"name":"QwQ 32B Preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "Qwen/Qwen2.5-72B-Instruct": {"name":"Qwen2.5 72B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "Qwen/Qwen2.5-Coder-32B-Instruct": {"name":"Qwen2.5 Coder 32B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "SD1.5": {"name":"SD1.5","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "SD1.5-ControlNet": {"name":"SD1.5-ControlNet","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "SD2": {"name":"SD2","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "SDXL-ControlNet": {"name":"SDXL-ControlNet","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "SDXL-turbo": {"name":"SDXL-turbo","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "SDXL1.0-base": {"name":"SDXL1.0-base","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "SSD": {"name":"SSD","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "deepseek-ai/DeepSeek-R1": {"name":"DeepSeek R1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "deepseek-ai/DeepSeek-R1-Zero": {"name":"DeepSeek R1 Zero","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "deepseek-ai/DeepSeek-V2.5": {"name":"DeepSeek V2.5","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Llama-3.2-3B-Instruct": {"name":"Llama 3.2 3B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Llama-3.3-70B-Instruct": {"name":"Llama 3.3 70B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Meta-Llama-3-70B-Instruct": {"name":"Llama 3 70B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Meta-Llama-3.1-70B-Instruct": {"name":"Llama 3.1 70B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Meta-Llama-3.1-8B-Instruct": {"name":"Llama 3.1 8B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}}
}},
"llamacpp": {"models": {
  "LLaMA_CPP": {"name":"llama.cpp server model","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}}
}},
"minimax": {"models": {
  "MiniMax-M2": {"name":"MiniMax-M2","description":"Efficient open MiniMax model built for coding agents and tool-heavy workflows","family":"minimax","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2},"release_date":"2025-10-27","last_updated":"2025-10-27","open_weights":true,"status":"deprecated"},
  "MiniMax-M2.1": {"name":"MiniMax-M2.1","description":"Earlier MiniMax agent model for practical coding and productivity tasks","family":"minimax","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03,"cache_write":0.375},"release_date":"2025-12-23","last_updated":"2025-12-23","open_weights":true,"status":"deprecated"},
  "MiniMax-M2.5": {"name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03,"cache_write":0.375},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"status":"deprecated"},
  "MiniMax-M2.5-highspeed": {"name":"MiniMax-M2.5-highspeed","description":"High-speed MiniMax model for low-latency coding and agent workflows","family":"minimax","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_read":0.03,"cache_write":0.375},"release_date":"2026-02-13","last_updated":"2026-02-13","open_weights":true,"status":"deprecated"},
  "MiniMax-M2.7": {"name":"MiniMax-M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06,"cache_write":0.375},"release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true,"status":"deprecated"},
  "MiniMax-M2.7-highspeed": {"name":"MiniMax-M2.7-highspeed","description":"Low-latency M2.7 variant for interactive coding plans and agent loops","family":"minimax","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_read":0.06,"cache_write":0.375},"release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true,"status":"deprecated"},
  "MiniMax-M3": {"name":"MiniMax-M3","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":524288},"cost":{"input":0.3,"output":1.2,"cache_read":0.06,"tiers":[{"input":0.6,"output":2.4,"cache_read":0.12,"tier":{"type":"context","size":512000}}],"context_over_200k":{"input":0.6,"output":2.4,"cache_read":0.12}},"release_date":"2026-06-01","last_updated":"2026-06-25","open_weights":true},
  "MiniMax-M3.1-Flash": {"name":"MiniMax-M3.1-Flash","description":"Frontier multimodal coding model with 1M context window and tunable thinking depth","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":524288},"cost":{"input":0.15,"output":0.6,"cache_read":0.015},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false},
  "image-01": {"name":"image-01","modalities":{"input":["text","image"]},"rig":{"pinned":["name","modalities"]}},
  "image-01-live": {"name":"image-01-live","modalities":{"input":["text","image"]},"rig":{"pinned":["name","modalities"]}},
  "speech-02-hd": {"name":"speech-02-hd","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "speech-02-turbo": {"name":"speech-02-turbo","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "speech-2.6-hd": {"name":"speech-2.6-hd","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "speech-2.6-turbo": {"name":"speech-2.6-turbo","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "speech-2.8-hd": {"name":"speech-2.8-hd","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "speech-2.8-turbo": {"name":"speech-2.8-turbo","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}}
}},
"mistral": {"models": {
  "codestral-2508": {"name":"Codestral 25.08","description":"Mistral code model for low-latency completion, fill-in-the-middle, and code generation","family":"codestral","canonical_model_id":"mistral/codestral-2508","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":256000,"output":8192},"cost":{"input":0.3,"output":0.9,"cache_read":0.03},"release_date":"2025-07-30","last_updated":"2025-07-30","open_weights":false},
  "codestral-embed": {"name":"Codestral Embed","modalities":{"input":["text"]},"limit":{"context":8192},"rig":{"pinned":["name","modalities","limit.context"]}},
  "codestral-latest": {"name":"Codestral (latest)","description":"Mistral code model for completions, refactors, and developer IDE workflows","family":"codestral","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":256000,"output":4096},"cost":{"input":0.3,"output":0.9,"cache_read":0.03},"knowledge":"2024-10","release_date":"2024-05-29","last_updated":"2025-01-04","open_weights":true,"rig":{"cache":["short"]}},
  "deepseek-v4-1-flash-1017": {"name":"DeepSeek V4.1 Flash 1017","description":"October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":128000},"cost":{"input":0.3,"output":1.2,"cache_read":0.007},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true},
  "devstral-2512": {"name":"Devstral 2","description":"Mistral's coding-agent model for repository work, terminal tasks, and software fixes","family":"devstral","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":2},"knowledge":"2025-12","release_date":"2025-12-09","last_updated":"2025-12-09","open_weights":true,"status":"deprecated"},
  "devstral-latest": {"name":"Devstral 2","description":"Legacy model retained for compatibility with older integrations","family":"devstral","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":2},"knowledge":"2025-12","release_date":"2025-12-09","last_updated":"2025-12-09","open_weights":true,"status":"deprecated"},
  "devstral-medium-latest": {"name":"Devstral 2 (latest)","description":"Legacy model retained for compatibility with older integrations","family":"devstral","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":2},"knowledge":"2025-12","release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true,"status":"deprecated"},
  "glm-5-2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.14},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true,"status":"deprecated"},
  "labs-devstral-small-2512": {"name":"Devstral Small 2","description":"Legacy model retained for compatibility with older integrations","family":"devstral","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":256000},"cost":{"input":0,"output":0},"knowledge":"2025-12","release_date":"2025-12-09","last_updated":"2025-12-09","open_weights":true,"status":"deprecated"},
  "labs-leanstral-1-5-1": {"name":"Leanstral 1.5","description":"An updated Lean 4 formal proof engineering model optimised for automated theorem proving and autoformalization. 119B total parameters, 6.5B active.","family":"leanstral","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":128000},"cost":{"input":0,"output":0},"release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":true,"status":"deprecated"},
  "magistral-medium-latest": {"name":"Magistral Medium (latest)","description":"Mistral reasoning model for transparent analysis, math, and complex decisions","family":"magistral-medium","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":2,"output":5},"knowledge":"2025-06","release_date":"2025-03-17","last_updated":"2025-03-20","open_weights":true,"status":"deprecated","rig":{"cache":["short"]}},
  "ministral-14b-2512": {"name":"Ministral 3 14B","description":"Open vision-language model for efficient local deployment, instruction following, and tool use","family":"ministral","canonical_model_id":"mistral/ministral-3-14b-instruct-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.2,"output":0.2,"cache_read":0.02},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "ministral-3b-2512": {"name":"Ministral 3 3B","description":"Compact open vision-language model for edge deployment, instruction following, and tool use","family":"ministral","canonical_model_id":"mistral/ministral-3-3b-instruct-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":131072,"output":262144},"cost":{"input":0.1,"output":0.1,"cache_read":0.01},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "ministral-3b-latest": {"name":"Ministral 3B (latest)","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":131072,"output":128000},"cost":{"input":0.1,"output":0.1,"cache_read":0.01},"knowledge":"2024-10","release_date":"2024-10-01","last_updated":"2024-10-04","open_weights":true},
  "ministral-8b-2512": {"name":"Ministral 3 8B","description":"Compact open vision-language model for edge deployment, instruction following, and tool use","family":"ministral","canonical_model_id":"mistral/ministral-3-8b-instruct-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.15,"output":0.15,"cache_read":0.015},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "ministral-8b-latest": {"name":"Ministral 8B (latest)","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":128000},"cost":{"input":0.15,"output":0.15,"cache_read":0.015},"knowledge":"2024-10","release_date":"2024-10-01","last_updated":"2024-10-04","open_weights":true},
  "mistral-embed": {"name":"Mistral Embed","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"mistral-embed","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8000,"output":3072},"cost":{"input":0.1,"output":0},"release_date":"2023-12-11","last_updated":"2023-12-11","open_weights":false},
  "mistral-large-2512": {"name":"Mistral Large 3","description":"Mistral's largest general model for enterprise agents, coding, and multilingual reasoning","family":"mistral-large","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":1.5,"cache_read":0.05},"knowledge":"2024-11","release_date":"2024-11-01","last_updated":"2025-12-02","open_weights":true},
  "mistral-large-4": {"name":"Mistral Large 4","description":"Mistral's 1T-parameter multimodal MoE for reasoning, coding, and agentic work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-4","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0.7,"output":2.1,"cache_read":0.07},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false,"status":"beta"},
  "mistral-large-4-0-8": {"name":"Mistral Large 4.0.8","description":"Mistral's 1T-parameter multimodal MoE for reasoning, coding, and agentic work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-4","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0.7,"output":2.1,"cache_read":0.07},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false,"status":"beta"},
  "mistral-large-latest": {"name":"Mistral Large (latest)","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":1.5,"cache_read":0.05},"knowledge":"2024-11","release_date":"2024-11-01","last_updated":"2025-12-02","open_weights":true,"rig":{"cache":["short"]}},
  "mistral-medium-2505": {"name":"Mistral Medium 3","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.4,"output":2},"knowledge":"2025-05","release_date":"2025-05-07","last_updated":"2025-05-07","open_weights":false,"status":"deprecated"},
  "mistral-medium-2508": {"name":"Mistral Medium 3.1","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":2},"knowledge":"2025-05","release_date":"2025-08-12","last_updated":"2025-08-12","open_weights":false,"status":"deprecated"},
  "mistral-medium-2604": {"name":"Mistral Medium 3.5","description":"Balanced Mistral model for enterprise assistants, multilingual work, and tools","family":"mistral-medium","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":1.5,"output":7.5,"cache_read":0.15},"release_date":"2026-04-29","last_updated":"2026-04-29","open_weights":true},
  "mistral-medium-latest": {"name":"Mistral Medium (latest)","description":"Balanced Mistral model for enterprise assistants, multilingual work, and tools","family":"mistral-medium","canonical_model_id":"mistral/mistral-medium-latest","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":1.5,"output":7.5,"cache_read":0.15},"release_date":"2026-04-29","last_updated":"2026-04-29","open_weights":true,"rig":{"cache":["short"]}},
  "mistral-small-2506": {"name":"Mistral Small 3.2","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.1,"output":0.3},"knowledge":"2025-03","release_date":"2025-06-20","last_updated":"2025-06-20","open_weights":true,"status":"deprecated"},
  "mistral-small-2603": {"name":"Mistral Small 4","description":"Fast Mistral production model for chat, extraction, and cost-sensitive agents","family":"mistral-small","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":256000},"cost":{"input":0.15,"output":0.6,"cache_read":0.015},"knowledge":"2025-06","release_date":"2026-03-16","last_updated":"2026-03-16","open_weights":true},
  "mistral-small-latest": {"name":"Mistral Small (latest)","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":256000},"cost":{"input":0.15,"output":0.6,"cache_read":0.015},"knowledge":"2025-06","release_date":"2026-03-16","last_updated":"2026-03-16","open_weights":true,"rig":{"cache":["short"]}},
  "open-mistral-nemo": {"name":"Open Mistral Nemo","description":"Legacy model retained for compatibility with older integrations","family":"mistral-nemo","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":128000},"cost":{"input":0.15,"output":0.15},"knowledge":"2024-07","release_date":"2024-07-01","last_updated":"2024-07-01","open_weights":true,"status":"deprecated"},
  "pixtral-12b": {"name":"Pixtral 12B","description":"Mistral vision-language model for image understanding and multimodal chat","family":"pixtral","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":128000},"cost":{"input":0.15,"output":0.15},"knowledge":"2024-09","release_date":"2024-09-01","last_updated":"2024-09-01","open_weights":true,"status":"deprecated"},
  "voxtral-mini-latest": {"name":"Voxtral Mini (latest)","description":"Speech transcription model for accurate audio-to-text and captioning workflows","family":"voxtral","type":"transcription","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["audio"],"output":["text"]},"limit":{"context":0,"output":0},"release_date":"2026-02-04","last_updated":"2026-02-04","open_weights":false},
  "voxtral-mini-tts-latest": {"name":"Voxtral Mini TTS (latest)","description":"Multilingual text-to-speech model with zero-shot voice cloning","family":"voxtral","type":"speech","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":0,"output":0},"release_date":"2026-03-23","last_updated":"2026-03-23","open_weights":true},
  "voxtral-small-2507": {"name":"Voxtral Small","description":"Instruct model with native audio input for speech understanding and tool use","family":"voxtral","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","audio","pdf"],"output":["text"]},"limit":{"context":32768,"output":32000},"cost":{"input":0.1,"output":0.4},"release_date":"2025-07-15","last_updated":"2025-07-15","open_weights":true},
  "voxtral-small-latest": {"name":"Voxtral Small (latest)","description":"Instruct model with native audio input for speech understanding and tool use","family":"voxtral","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","audio","pdf"],"output":["text"]},"limit":{"context":32768,"output":32000},"cost":{"input":0.1,"output":0.4},"release_date":"2025-07-15","last_updated":"2025-07-15","open_weights":true},
  "zai-glm-5-2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.14},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true,"status":"deprecated"},
  "zai-glm-5-3": {"name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.14},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "zai-glm-5-3-flash": {"name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true}
}},
"moonshot": {"models": {
  "kimi-k2-0711-preview": {"name":"kimi-k2-0711-preview","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "kimi-k2-0905-preview": {"name":"kimi-k2-0905-preview","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "kimi-k2-thinking": {"name":"kimi-k2-thinking","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "kimi-k2-turbo-preview": {"name":"kimi-k2-turbo-preview","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "kimi-k2.6": {"name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.6","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.16},"knowledge":"2025-01","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true},
  "kimi-k2.7-code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true},
  "kimi-k2.7-code-highspeed": {"name":"Kimi K2.7 Code HighSpeed","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":1.9,"output":8,"cache_read":0.38},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true},
  "kimi-k3": {"name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":1048576},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true,"rig":{"reasoning_field":"reasoning_content"}},
  "kimi-k3.1-pro": {"name":"Kimi K3.1 Pro","description":"Multimodal Kimi model with 1M context and always-on reasoning for coding, 3D creation, financial research, and agentic work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3.1-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":1048576},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":false},
  "kimi-k3.1-pro-highspeed": {"name":"Kimi K3.1 Pro Highspeed","description":"Multimodal Kimi model with 1M context and always-on reasoning for coding, 3D creation, financial research, and agentic work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3.1-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":1048576},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":4},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":false},
  "moonshot-v1-128k": {"name":"moonshot-v1-128k","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "moonshot-v1-128k-vision-preview": {"name":"moonshot-v1-128k vision","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "moonshot-v1-32k": {"name":"moonshot-v1-32k","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "moonshot-v1-32k-vision-preview": {"name":"moonshot-v1-32k vision","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "moonshot-v1-8k": {"name":"moonshot-v1-8k","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "moonshot-v1-8k-vision-preview": {"name":"moonshot-v1-8k vision","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}}
}},
"ollama": {"models": {
  "all-minilm": {"name":"all-minilm","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "beam": {"name":"Beam 1.0 501B A23B","description":"Sparse mixture-of-experts reasoning model with 23B active parameters for coding and agentic workflows","family":"beam","canonical_model_id":"reflection/beam-1.0-501b-a23b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"knowledge":"2026-06-30","release_date":"2026-10-05","last_updated":"2026-10-16","open_weights":true},
  "bge-m3": {"name":"bge-m3","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "deepseek-r1": {"name":"DeepSeek R1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "deepseek-v3.1": {"name":"DeepSeek V3.1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "deepseek-v4-pro": {"name":"deepseek-v4-pro","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":1048576},"cost":{"input":0.66,"output":1.98,"cache_read":0.022},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true},
  "deepseek-v4-pro:0813": {"name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro-0813","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":1048576},"cost":{"input":0.66,"output":1.98,"cache_read":0.022},"release_date":"2026-08-13","last_updated":"2026-08-22","open_weights":true},
  "deepseek-v4.1-flash": {"name":"DeepSeek V4.1 Flash","description":"DeepSeek V4.1 Flash model for reasoning and agentic coding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.15,"output":0.6,"cache_read":0.003},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true},
  "deepseek-v4.1-flash:1017": {"name":"DeepSeek V4.1 Flash 1017","description":"October 17 update of DeepSeek V4.1 Flash with improved reasoning, coding, agentic tasks, and visual understanding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.15,"output":0.6,"cache_read":0.003},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true},
  "embeddinggemma": {"name":"embeddinggemma","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "gemma3": {"name":"Gemma 3","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "gemma4": {"name":"Gemma 4","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "gemma4:31b": {"name":"gemma4:31b","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-31b-it","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.14,"output":0.4,"cache_read":0.05},"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true},
  "glm-5.2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":976000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true},
  "glm-5.3": {"name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "glm-5.3-flash": {"name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true},
  "gpt-oss": {"name":"gpt-oss","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "gpt-oss:120b": {"name":"gpt-oss:120b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.15,"output":0.6,"cache_read":0.014},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "gpt-oss:20b": {"name":"gpt-oss:20b","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.07,"output":0.3,"cache_read":0.035},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "kimi-k2.6": {"name":"kimi-k2.6","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.6","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.16},"knowledge":"2025-01","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true},
  "kimi-k2.7-code": {"name":"kimi-k2.7-code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true},
  "kimi-k3": {"name":"kimi-k3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "llama3.1": {"name":"Llama 3.1","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "llama3.2": {"name":"Llama 3.2","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "llama3.3": {"name":"Llama 3.3","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "llama4": {"name":"Llama 4","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "llava": {"name":"LLaVA","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "minimax-m2.7": {"name":"minimax-m2.7","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.7","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":196608,"output":196608},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true},
  "minimax-m3": {"name":"minimax-m3","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M3","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":512000,"output":131072},"cost":{"input":0.6,"output":2.4,"cache_read":0.12},"knowledge":"2025-01","release_date":"2026-06-01","last_updated":"2026-06-01","open_weights":true},
  "minimax-m3.1-flash": {"name":"MiniMax-M3.1-Flash","description":"Frontier multimodal coding model with 1M context window and tunable thinking depth","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":512000},"cost":{"input":0.15,"output":0.6,"cache_read":0.015},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false},
  "mistral": {"name":"Mistral","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "mistral-large-3:675b": {"name":"mistral-large-3:675b","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-2512","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":1.5},"knowledge":"2024-11","release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistral-large-4": {"name":"Mistral Large 4","description":"Mistral's 1T-parameter multimodal MoE for reasoning, coding, and agentic work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-4","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0.68,"output":2.09,"cache_read":0.07},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false},
  "mistral-small3.2": {"name":"Mistral Small 3.2","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "muse-spark-open": {"name":"Muse Spark Open","description":"Muse Spark Open is an 816B total / 42B active open-weight multimodal reasoning model from Meta for coding and long-horizon agentic work, with 1M-token context.","family":"muse","canonical_model_id":"meta/muse-spark-open","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"knowledge":"2026-01-04","release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true},
  "mxbai-embed-large": {"name":"mxbai-embed-large","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "nemotron-3-nano:30b": {"name":"nemotron-3-nano:30b","description":"Small Nemotron 3 MoE for efficient coding, math, and long-context agents","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-nano-30b-a3b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.06,"output":0.24},"knowledge":"2025-06-25","release_date":"2025-12-15","last_updated":"2025-12-15","open_weights":true},
  "nemotron-3-super": {"name":"nemotron-3-super","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-super-120b-a12b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.015,"output":0.6,"cache_read":0.015},"knowledge":"2025-06","release_date":"2026-03-11","last_updated":"2026-03-11","open_weights":true},
  "nemotron-3-ultra": {"name":"nemotron-3-ultra","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-ultra-550b-a55b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":128000},"cost":{"input":0.1,"output":3,"cache_read":0.1},"knowledge":"2025-09","release_date":"2026-06-04","last_updated":"2026-06-04","open_weights":true},
  "nomic-embed-text": {"name":"nomic-embed-text","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "phi4": {"name":"Phi-4","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "qwen3": {"name":"Qwen3","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "qwen3-coder": {"name":"Qwen3 Coder","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "qwen3-embedding": {"name":"qwen3-embedding","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "qwen3.5": {"name":"Qwen3.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "qwen3.6": {"name":"Qwen3.6","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "qwen3.8": {"name":"Qwen3.8","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}}
}},
"openai": {"models": {
  "chatgpt-image-latest": {"name":"chatgpt-image-latest","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":0,"input":0,"output":0},"release_date":"2025-12-16","last_updated":"2025-12-16","open_weights":false,"status":"deprecated"},
  "dall-e-2": {"name":"DALL-E 2","modalities":{"input":["text","image"]},"rig":{"pinned":["name","modalities"]}},
  "dall-e-3": {"name":"DALL-E 3","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "gpt-3.5-turbo": {"name":"GPT-3.5-turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":16385,"output":4096},"cost":{"input":0.5,"output":1.5,"cache_read":0},"knowledge":"2021-09-01","release_date":"2023-03-01","last_updated":"2023-11-06","open_weights":false,"status":"deprecated","rig":{"cache":["short"]}},
  "gpt-4": {"name":"GPT-4","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":8192},"cost":{"input":30,"output":60},"knowledge":"2023-11","release_date":"2023-11-06","last_updated":"2024-04-09","open_weights":false,"status":"deprecated","rig":{"cache":["short"]}},
  "gpt-4-0125-preview": {"name":"GPT-4 Turbo preview (gpt-4-0125-preview)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":128000,"output":4096},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4-0613": {"name":"GPT-4 (0613)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":8192,"output":8192},"cost":{"input":30,"output":60},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","cost.input","cost.output","status"]}},
  "gpt-4-1106-preview": {"name":"GPT-4 Turbo preview (gpt-4-1106-preview)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":128000,"output":4096},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4-1106-vision-preview": {"name":"GPT-4 Turbo with vision (gpt-4-1106-vision-preview)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":4096},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4-32k": {"name":"GPT-4 32K","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":32768,"output":32768},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4-32k-0613": {"name":"GPT-4 32K (0613)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":32768,"output":32768},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4-turbo": {"name":"GPT-4 Turbo","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":10,"output":30},"knowledge":"2023-12","release_date":"2023-11-06","last_updated":"2024-04-09","open_weights":false,"status":"deprecated","rig":{"cache":["short"]}},
  "gpt-4-turbo-2024-04-09": {"name":"GPT-4 Turbo (2024-04-09)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":4096},"cost":{"input":10,"output":30},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","cost.input","cost.output","status"]}},
  "gpt-4-turbo-preview": {"name":"GPT-4 Turbo preview (gpt-4-turbo-preview)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":128000,"output":4096},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4-vision-preview": {"name":"GPT-4 Turbo with vision (gpt-4-vision-preview)","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":4096},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4.1": {"name":"GPT-4.1","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":3.5,"output":14,"cache_read":0.875},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short","long"]}},
  "gpt-4.1-2025-04-14": {"name":"GPT-4.1","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":3.5,"output":14,"cache_read":0.875},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short","long"],"pinned":["name","description","family","reasoning","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","knowledge","release_date","last_updated","open_weights","experimental"]}},
  "gpt-4.1-mini": {"name":"GPT-4.1 mini","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":0.7,"output":2.8,"cache_read":0.175},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short","long"]}},
  "gpt-4.1-nano": {"name":"GPT-4.1 nano","description":"Tiny GPT-4.1 option for classification, routing, and very high-volume tasks","family":"gpt-nano","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":0.1,"output":0.4,"cache_read":0.025},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false,"status":"deprecated","experimental":{"modes":{"fast":{"cost":{"input":0.2,"output":0.8,"cache_read":0.05},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short","long"]}},
  "gpt-4.5-preview": {"name":"GPT-4.5 preview","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":16384},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4.5-preview-2025-02-27": {"name":"GPT-4.5 preview","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":16384},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","temperature","modalities","limit.context","limit.output","status"]}},
  "gpt-4o": {"name":"GPT-4o","description":"Omni-era GPT for multimodal chat, practical coding, and general assistants","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4.25,"output":17,"cache_read":2.125},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short"]}},
  "gpt-4o-2024-05-13": {"name":"GPT-4o (2024-05-13)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":5,"output":15},"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-05-13","open_weights":false,"status":"deprecated","experimental":{"modes":{"fast":{"cost":{"input":8.75,"output":26.25},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short"]}},
  "gpt-4o-2024-08-06": {"name":"GPT-4o (2024-08-06)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-08-06","last_updated":"2024-08-06","open_weights":false,"rig":{"cache":["short"]}},
  "gpt-4o-2024-11-20": {"name":"GPT-4o (2024-11-20)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-11-20","last_updated":"2024-11-20","open_weights":false,"rig":{"cache":["short"]}},
  "gpt-4o-mini": {"name":"GPT-4o mini","description":"Small omni GPT for cheap multimodal assistance and production-scale traffic","family":"gpt-mini","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075},"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":0.25,"output":1,"cache_read":0.125},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short"]}},
  "gpt-5": {"name":"GPT-5","description":"Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows","family":"gpt","canonical_model_id":"openai/gpt-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":2.5,"output":20,"cache_read":0.25},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":0.625,"output":5,"cache_read":0.0625},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short","long"],"sampling":"never"}},
  "gpt-5-mini": {"name":"GPT-5 Mini","description":"Small GPT-5 for responsive agents, coding help, and everyday automation","family":"gpt-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":0.45,"output":3.6,"cache_read":0.045},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":0.125,"output":1,"cache_read":0.0125},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short","long"],"sampling":"never"}},
  "gpt-5-nano": {"name":"GPT-5 Nano","description":"Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs","family":"gpt-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"experimental":{"modes":{"flex":{"cost":{"input":0.025,"output":0.2,"cache_read":0.0025},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short","long"],"sampling":"never"}},
  "gpt-5-pro": {"name":"GPT-5 Pro","description":"Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning","family":"gpt-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":272000},"cost":{"input":15,"output":120},"knowledge":"2024-09-30","release_date":"2025-10-06","last_updated":"2025-10-06","open_weights":false,"rig":{"cache":["short","long"],"sampling":"never"}},
  "gpt-5.1": {"name":"GPT-5.1","description":"Sharper GPT-5 generation for coding, product work, and tool-assisted tasks","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":2.5,"output":20,"cache_read":0.25},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":0.625,"output":5,"cache_read":0.0625},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.2": {"name":"GPT-5.2","description":"Reliable GPT generation for broad coding, writing, and tool-assisted product work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":3.5,"output":28,"cache_read":0.35},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":0.875,"output":7,"cache_read":0.0875},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.2-chat-latest": {"name":"GPT-5.2 Chat","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium"]}],"tool_call":true,"structured_output":true,"temperature":false,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":16384},"cost":{"input":1.75,"output":14,"cache_read":0.175},"status":"deprecated","rig":{"cache":["short","long"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","status"],"sampling":"reasoning_off"}},
  "gpt-5.2-pro": {"name":"GPT-5.2 Pro","description":"Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows","family":"gpt-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":21,"output":168},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.3-chat-latest": {"name":"GPT-5.3 Chat (latest)","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":128000,"output":16384},"cost":{"input":1.75,"output":14,"cache_read":0.175},"status":"deprecated","rig":{"cache":["short","long"],"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","status"],"sampling":"reasoning_off"}},
  "gpt-5.3-codex": {"name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":3.5,"output":28,"cache_read":0.35},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.3-codex-spark": {"name":"GPT-5.3 Codex Spark","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex-spark","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"input":100000,"output":32000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"rig":{"cache":["short","long"],"pinned":["modalities"],"sampling":"reasoning_off"}},
  "gpt-5.4": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":5,"output":30,"cache_read":0.5},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.4-mini": {"name":"GPT-5.4 mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":1.5,"output":9,"cache_read":0.15},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":0.375,"output":2.25,"cache_read":0.0375},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.4-nano": {"name":"GPT-5.4 nano","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"experimental":{"modes":{"flex":{"cost":{"input":0.1,"output":0.625,"cache_read":0.01},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.4-pro": {"name":"GPT-5.4 Pro","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"rig":{"cache":["short","long"],"sampling":"reasoning_off"}},
  "gpt-5.5": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":12.5,"output":75,"cache_read":1.25},"provider":{"body":{"service_tier":"priority"}}}}},"rig":{"cache":["long"]}},
  "gpt-5.5-pro": {"name":"GPT-5.5 Pro","description":"Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding","family":"gpt-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"rig":{"cache":["long"]}},
  "gpt-5.6": {"name":"GPT-5.6","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":8,"output":40,"cache_read":0.8,"cache_write":10},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["none","short","long"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":0.4,"output":2.4,"cache_read":0.04,"cache_write":0.5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["none","short","long"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-sol": {"name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":8,"output":40,"cache_read":0.8,"cache_write":10},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["none","short","long"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-terra": {"name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4,"output":24,"cache_read":0.4,"cache_write":5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["none","short","long"],"prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":20,"output":100,"cache_read":2,"cache_write":25},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}},"ultrafast":{"cost":{"input":60,"output":300,"cache_read":6,"cache_write":75},"provider":{"body":{"service_tier":"ultrafast"}}}}},"rig":{"cache":["none","short","long"],"chat_tools":"never","prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-luna": {"name":"GPT-6 Luna","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":0.2,"output":1,"cache_read":0.02,"cache_write":0.25},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["none","short","long"],"chat_tools":"reasoning_off","prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-sol": {"name":"GPT-6 Sol","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["none","short","long"],"chat_tools":"reasoning_off","prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-sol": {"name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}},"ultrafast":{"cost":{"input":12,"output":60,"cache_read":0.6,"cache_write":15},"provider":{"body":{"service_tier":"ultrafast"}}}}},"rig":{"cache":["none","short","long"],"chat_tools":"never","prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-terra": {"name":"GPT-6.1 Terra","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"provider":{"body":{"service_tier":"priority"}}},"pro":{"provider":{"body":{"reasoning":{"mode":"pro"}}}}}},"rig":{"cache":["none","short","long"],"chat_tools":"never","prompt_cache_options":true,"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-daybreak-blue-latest": {"name":"Daybreak Blue","description":"Alias for flagship general-purpose models with safeguards for defensive cybersecurity work","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-08-07","last_updated":"2026-08-07","open_weights":false},
  "gpt-daybreak-red-latest": {"name":"Daybreak Red","description":"Alias for advanced cybersecurity models for authorized vulnerability research and security testing","family":"gpt","canonical_model_id":"openai/gpt-5.6-cyber","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":12.5,"output":75,"cache_read":1.25,"cache_write":15.625},"knowledge":"2026-02-16","release_date":"2026-08-07","last_updated":"2026-08-07","open_weights":false},
  "gpt-image-1-mini": {"name":"gpt-image-1-mini","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":0,"input":0,"output":0},"release_date":"2025-09-26","last_updated":"2025-09-26","open_weights":false,"status":"deprecated"},
  "gpt-image-1.5": {"name":"gpt-image-1.5","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","canonical_model_id":"openai/gpt-image-1.5","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":0,"input":0,"output":0},"release_date":"2025-11-25","last_updated":"2025-11-25","open_weights":false,"status":"deprecated"},
  "gpt-image-2": {"name":"gpt-image-2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt-image","type":"image","canonical_model_id":"openai/gpt-image-2","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["image"]},"limit":{"context":0,"input":0,"output":0},"cost":{"input":5,"output":30,"cache_read":1.25},"release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false},
  "gpt-realtime-2.1": {"name":"GPT-Realtime-2.1","description":"Realtime speech-to-speech model with configurable reasoning, tool use, and robust voice-agent behavior","family":"gpt","type":"realtime","canonical_model_id":"openai/gpt-realtime-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","audio","image"],"output":["text","audio"]},"limit":{"context":128000,"input":96000,"output":32000},"cost":{"input":4,"output":24,"cache_read":0.4,"input_audio":32,"output_audio":64},"knowledge":"2024-09-30","release_date":"2026-07-06","last_updated":"2026-07-06","open_weights":false},
  "o1": {"name":"o1","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5},"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","open_weights":false,"status":"deprecated","rig":{"cache":["short"],"sampling":"never"}},
  "o1-2024-12-17": {"name":"o1","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":15,"output":60,"cache_read":7.5},"knowledge":"2023-09","release_date":"2024-12-05","last_updated":"2024-12-05","open_weights":false,"status":"deprecated","rig":{"cache":["short"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","knowledge","release_date","last_updated","open_weights","status"],"sampling":"never"}},
  "o1-mini": {"name":"o1-mini","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":false,"modalities":{"input":["text"]},"limit":{"context":128000,"output":65536},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities","limit.context","limit.output","status"],"reasoning_control":"none","sampling":"never"}},
  "o1-mini-2024-09-12": {"name":"o1-mini","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":false,"modalities":{"input":["text"]},"limit":{"context":128000,"output":65536},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities","limit.context","limit.output","status"],"reasoning_control":"none","sampling":"never"}},
  "o1-preview": {"name":"o1-preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":false,"modalities":{"input":["text"]},"limit":{"context":128000,"output":32768},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities","limit.context","limit.output","status"],"reasoning_control":"none","sampling":"never"}},
  "o1-preview-2024-09-12": {"name":"o1-preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":false,"modalities":{"input":["text"]},"limit":{"context":128000,"output":32768},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities","limit.context","limit.output","status"],"reasoning_control":"none","sampling":"never"}},
  "o1-pro": {"name":"o1-pro","description":"O-series reasoning model for hard analysis, math, coding, and planning","family":"o-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":150,"output":600},"knowledge":"2023-09","release_date":"2025-03-19","last_updated":"2025-03-19","open_weights":false,"status":"deprecated","rig":{"cache":["short"],"sampling":"never"}},
  "o3": {"name":"o3","description":"Deliberate o-series reasoner for hard math, coding, and multi-step analysis","family":"o","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":0.5},"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","open_weights":false,"experimental":{"modes":{"fast":{"cost":{"input":3.5,"output":14,"cache_read":0.875},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":1,"output":4,"cache_read":0.25},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short"],"sampling":"never"}},
  "o3-mini": {"name":"o3-mini","description":"Smaller o-series reasoner for economical coding, math, and planning tasks","family":"o-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55},"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","open_weights":false,"status":"deprecated","rig":{"cache":["short"],"sampling":"never"}},
  "o3-mini-2025-01-31": {"name":"o3-mini","description":"Smaller o-series reasoner for economical coding, math, and planning tasks","family":"o-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.55},"knowledge":"2024-05","release_date":"2024-12-20","last_updated":"2025-01-29","open_weights":false,"status":"deprecated","rig":{"cache":["short"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","knowledge","release_date","last_updated","open_weights","status"],"sampling":"never"}},
  "o3-pro": {"name":"o3-pro","description":"High-effort o3 tier for difficult technical reasoning and careful answers","family":"o-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":20,"output":80},"knowledge":"2024-05","release_date":"2025-06-10","last_updated":"2025-06-10","open_weights":false,"rig":{"cache":["short"],"sampling":"never"}},
  "o4-mini": {"name":"o4-mini","description":"Fast o-series model for compact reasoning, coding, and tool use","family":"o-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.275},"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","open_weights":false,"status":"deprecated","experimental":{"modes":{"fast":{"cost":{"input":2,"output":8,"cache_read":0.5},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":0.55,"output":2.2,"cache_read":0.1375},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short"],"sampling":"never"}},
  "o4-mini-2025-04-16": {"name":"o4-mini","description":"Fast o-series model for compact reasoning, coding, and tool use","family":"o-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":1.1,"output":4.4,"cache_read":0.275},"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","open_weights":false,"status":"deprecated","experimental":{"modes":{"fast":{"cost":{"input":2,"output":8,"cache_read":0.5},"provider":{"body":{"service_tier":"priority"}}},"flex":{"cost":{"input":0.55,"output":2.2,"cache_read":0.1375},"provider":{"body":{"service_tier":"flex"}}}}},"rig":{"cache":["short"],"pinned":["name","description","family","reasoning","reasoning_options","tool_call","structured_output","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","cost.cache_read","knowledge","release_date","last_updated","open_weights","status","experimental"],"sampling":"never"}},
  "text-embedding-3-large": {"name":"text-embedding-3-large","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8191,"output":3072},"cost":{"input":0.13,"output":0},"knowledge":"2024-01","release_date":"2024-01-25","last_updated":"2024-01-25","open_weights":false},
  "text-embedding-3-small": {"name":"text-embedding-3-small","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8191,"output":1536},"cost":{"input":0.02,"output":0},"knowledge":"2024-01","release_date":"2024-01-25","last_updated":"2024-01-25","open_weights":false},
  "text-embedding-ada-002": {"name":"text-embedding-ada-002","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"text-embedding","type":"embedding","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":1536},"cost":{"input":0.1,"output":0},"knowledge":"2022-12","release_date":"2022-12-15","last_updated":"2022-12-15","open_weights":false},
  "tts-1": {"name":"TTS-1","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "tts-1-hd": {"name":"TTS-1 HD","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "whisper-1": {"name":"Whisper","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}}
}},
"opencode": {"models": {
  "beam-free": {"name":"Beam Free","description":"Sparse mixture-of-experts reasoning model with 23B active parameters for coding and agentic workflows","family":"beam","canonical_model_id":"reflection/beam-1.0-501b-a23b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2026-06-30","release_date":"2026-10-05","last_updated":"2026-10-16","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "big-pickle": {"name":"Big Pickle","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"big-pickle","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"input":160000,"output":32000},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0},"knowledge":"2025-01","release_date":"2025-10-17","last_updated":"2025-10-17","open_weights":false,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "cat-jam-free": {"name":"Cat Jam Free","description":"A text-only reasoning model for coding and agentic workflows, with tool calling and structured JSON outputs.","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-10-12","last_updated":"2026-10-12","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "claude-3-5-haiku": {"name":"Claude Haiku 3.5","description":"Legacy model retained for compatibility with older integrations","family":"claude-haiku","canonical_model_id":"anthropic/claude-3-5-haiku-20241022","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":8192},"cost":{"input":0.8,"output":4,"cache_read":0.08,"cache_write":1},"knowledge":"2024-07-31","release_date":"2024-10-22","last_updated":"2024-10-22","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "claude-fable-5": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"sampling":"never"}},
  "claude-fable-5-1": {"name":"Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never"}},
  "claude-haiku-4-5": {"name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"cache":["none","short","long"],"prefill":true}},
  "claude-haiku-5-5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "claude-opus-4-1": {"name":"Claude Opus 4.1","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-1","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"cache":["none","short","long"],"prefill":true}},
  "claude-opus-4-5": {"name":"Claude Opus 4.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"cache":["none","short","long"],"prefill":true}},
  "claude-opus-4-6": {"name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"]}},
  "claude-opus-4-7": {"name":"Claude Opus 4.7","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"sampling":"never"}},
  "claude-opus-4-8": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"sampling":"never"}},
  "claude-opus-5": {"name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"sampling":"never"}},
  "claude-opus-5-5": {"name":"Claude Opus 5.5","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "claude-sonnet-4": {"name":"Claude Sonnet 4","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-0","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}},"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"cache":["none","short","long"],"prefill":true}},
  "claude-sonnet-4-5": {"name":"Claude Sonnet 4.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"cache":["none","short","long"],"prefill":true}},
  "claude-sonnet-4-6": {"name":"Claude Sonnet 4.6","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"]}},
  "claude-sonnet-5": {"name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"sampling":"never"}},
  "claude-sonnet-5-5": {"name":"Claude Sonnet 5.5","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "deepseek-v4-flash": {"name":"DeepSeek V4 Flash","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-0731","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.028},"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "deepseek-v4-flash-free": {"name":"DeepSeek V4 Flash Free","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-0731","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":128000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "deepseek-v4-flash-vision-exp": {"name":"DeepSeek V4 Flash Vision Exp","description":"Experimental multimodal DeepSeek V4 Flash model for image understanding, coding, and agentic work","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-vision-exp","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.028},"release_date":"2026-08-21","last_updated":"2026-09-01","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "deepseek-v4-pro": {"name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":1.74,"output":3.48,"cache_read":0.145},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "deepseek-v4.1-flash": {"name":"DeepSeek V4.1 Flash","description":"DeepSeek V4.1 Flash model for reasoning and agentic coding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":0.3,"output":1.2,"cache_read":0.006},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "deepseek-v4.1-flash-free": {"name":"DeepSeek V4.1 Flash Free","description":"DeepSeek V4.1 Flash model for reasoning and agentic coding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "exo-free": {"name":"Exo Free","description":"Free-tier preview reasoning model with text and image input and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["high"]}],"tool_call":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "fledge-alpha-free": {"name":"Fledge Alpha Free","description":"Free-tier preview reasoning model with text and image input and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0},"release_date":"2026-10-01","last_updated":"2026-10-01","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "flint-40b-free": {"name":"Flint 40B Free","description":"Open-weight mixture-of-experts model with about 2.5B active parameters for local coding agents","family":"flint","canonical_model_id":"recursive/flint","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2026-06","release_date":"2026-10-21","last_updated":"2026-10-21","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-4.6": {"name":"GLM-4.6","description":"Legacy model retained for compatibility with older integrations","family":"glm","canonical_model_id":"zhipuai/glm-4.6","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.1},"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-4.7": {"name":"GLM-4.7","description":"Legacy model retained for compatibility with older integrations","family":"glm","canonical_model_id":"zhipuai/glm-4.7","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.1},"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-4.7-free": {"name":"GLM-4.7 Free","description":"Legacy model retained for compatibility with older integrations","family":"glm","canonical_model_id":"zhipuai/glm-4.7","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-5": {"name":"GLM-5","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2},"knowledge":"2025-04","release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-5-free": {"name":"GLM-5 Free","description":"Legacy model retained for compatibility with older integrations","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-04","release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-5.1": {"name":"GLM-5.1","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-5.1","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"knowledge":"2025-04","release_date":"2026-04-07","last_updated":"2026-04-07","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-5.2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-5.3": {"name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-5.3-flash": {"name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "glm-5.3-flash-free": {"name":"GLM-5.3-Flash Free","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5": {"name":"GPT-5","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.07,"output":8.5,"cache_read":0.107},"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"never"}},
  "gpt-5-codex": {"name":"GPT-5 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.07,"output":8.5,"cache_read":0.107},"knowledge":"2024-09-30","release_date":"2025-09-15","last_updated":"2025-09-15","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5-nano": {"name":"GPT-5 Nano","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-nano","canonical_model_id":"openai/gpt-5-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"never"}},
  "gpt-5.1": {"name":"GPT-5.1","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-5.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.07,"output":8.5,"cache_read":0.107},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.1-codex": {"name":"GPT-5.1 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.1-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.07,"output":8.5,"cache_read":0.107},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5.1-codex-max": {"name":"GPT-5.1 Codex Max","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.1-codex-max","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5.1-codex-mini": {"name":"GPT-5.1 Codex Mini","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.1-codex-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5.2": {"name":"GPT-5.2","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.2-codex": {"name":"GPT-5.2 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.2-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5.3-codex": {"name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.3-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.3-codex-spark": {"name":"GPT-5.3 Codex Spark","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex-spark","canonical_model_id":"openai/gpt-5.3-codex-spark","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"input":128000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","canonical_model_id":"openai/gpt-5.4","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-mini": {"name":"GPT-5.4 Mini","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","canonical_model_id":"openai/gpt-5.4-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-nano": {"name":"GPT-5.4 Nano","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-nano","canonical_model_id":"openai/gpt-5.4-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.4-pro": {"name":"GPT-5.4 Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-pro","canonical_model_id":"openai/gpt-5.4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"cache_read":30},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"sampling":"reasoning_off"}},
  "gpt-5.5": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","canonical_model_id":"openai/gpt-5.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5.5-pro": {"name":"GPT-5.5 Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-pro","canonical_model_id":"openai/gpt-5.5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"cache_read":30},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-sol": {"name":"GPT-5.6 Sol","description":"Frontier GPT-5.6 model for complex professional work, coding, and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-5.6-terra": {"name":"GPT-5.6 Terra","description":"Balanced GPT-5.6 model for capable, cost-efficient everyday work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT-6 Astra is OpenAI's most capable model for complex reasoning, coding, computer use, research, and document creation.","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-luna": {"name":"GPT-6 Luna","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-sol": {"name":"GPT-6 Sol","description":"OpenAI model for complex coding and agentic workflows","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6.1-sol": {"name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "grok-4.5": {"name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.3,"tiers":[{"input":4,"output":12,"cache_read":0.6,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":0.6}},"release_date":"2026-07-08","last_updated":"2026-07-08","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "grok-4.6": {"name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "grok-4.7": {"name":"Grok 4.7","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "grok-build-0.1": {"name":"Grok Build 0.1","description":"Fast Grok coding model tuned for agentic engineering and iterative edits","family":"grok-build","canonical_model_id":"xai/grok-build-0.1","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":256000,"output":256000},"cost":{"input":1,"output":2,"cache_read":0.2},"release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "grok-code": {"name":"Grok Code Fast 1","description":"Legacy model retained for compatibility with older integrations","family":"grok","canonical_model_id":"xai/grok-code-fast-1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":256000},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0},"release_date":"2025-09-02","last_updated":"2025-09-02","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "hy3-free": {"name":"Hy3 Free","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy3","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":190000,"input":192000,"output":64000},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-07-06","last_updated":"2026-07-06","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "hy3-preview-free": {"name":"Hy3 preview Free","description":"Legacy model retained for compatibility with older integrations","family":"Hy","canonical_model_id":"tencent/hy3-preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-06","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "kimi-k2": {"name":"Kimi K2","description":"Legacy model retained for compatibility with older integrations","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2-0905","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":2.5,"cache_read":0.4},"knowledge":"2024-10","release_date":"2025-09-05","last_updated":"2025-09-05","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "kimi-k2-thinking": {"name":"Kimi K2 Thinking","description":"Legacy model retained for compatibility with older integrations","family":"kimi-thinking","canonical_model_id":"moonshotai/kimi-k2-thinking","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.4,"output":2.5,"cache_read":0.4},"knowledge":"2024-08","release_date":"2025-11-06","last_updated":"2025-11-06","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "kimi-k2.5": {"name":"Kimi K2.5","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.6,"output":3,"cache_read":0.1},"knowledge":"2025-01","release_date":"2026-01-27","last_updated":"2026-01-27","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "kimi-k2.5-free": {"name":"Kimi K2.5 Free","description":"Legacy model retained for compatibility with older integrations","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-01","release_date":"2026-01-27","last_updated":"2026-01-27","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "kimi-k2.6": {"name":"Kimi K2.6","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.6","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.95,"output":4,"cache_read":0.16},"knowledge":"2025-01","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "kimi-k2.7-code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "kimi-k3": {"name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "laguna-s-2.1-free": {"name":"Laguna S 2.1 Free","description":"Agentic coding model from Poolside in the XS size class for local deployment","family":"laguna","canonical_model_id":"poolside/laguna-s-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":32000},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "ling-2.6-flash-free": {"name":"Ling 2.6 Flash Free","description":"Legacy model retained for compatibility with older integrations","family":"ling-flash-free","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262100,"output":32800},"cost":{"input":0,"output":0},"knowledge":"2025-06","release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "ling-3.0-flash-fin-free": {"name":"Ling 3.0 Flash Fin Free","description":"Finance-enhanced model for financial research, multi-step investment workflows, and long-horizon planning and execution","family":"ling","canonical_model_id":"inclusionai/ling-3.0-flash-fin","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-08-27","last_updated":"2026-09-03","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "ling-3.0-flash-free": {"name":"Ling-3.0-flash Free","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","canonical_model_id":"inclusionai/ling-3.0-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-07-23","last_updated":"2026-07-23","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "ling-3.0-tiny-free": {"name":"Ling-3.0-tiny Free","description":"Compact MoE model for responsive agents, instruction following, and multi-turn conversations","family":"ling","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0},"release_date":"2026-08-06","last_updated":"2026-08-06","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "ling-3.1-flash-free": {"name":"Ling 3.1 Flash Free","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","canonical_model_id":"inclusionai/ling-3.1-flash","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "ll-alpha-free": {"name":"LL Alpha Free","description":"Best in class local model","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2026-06","release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "longcat-2.0-free": {"name":"LongCat-2.0 Free","description":"Meituan LongCat-2.0, a reasoning model with tool calling and a 1M-token context window","family":"longcat","canonical_model_id":"meituan/longcat-2.0","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "longcat-2.5-preview-free": {"name":"LongCat 2.5 Preview Free","description":"Meituan's multimodal reasoning model for coding and agent tasks, with image understanding and a 1M-token context window","family":"longcat","canonical_model_id":"meituan/longcat-2.5-preview","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-09-25","last_updated":"2026-09-25","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "mimo-v2-flash-free": {"name":"MiMo V2 Flash Free","description":"Legacy model retained for compatibility with older integrations","family":"mimo","canonical_model_id":"xiaomi/mimo-v2-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2024-12-01","release_date":"2025-12-16","last_updated":"2026-02-04","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "mimo-v2-omni-free": {"name":"MiMo V2 Omni Free","description":"Legacy model retained for compatibility with older integrations","family":"mimo","canonical_model_id":"xiaomi/mimo-v2-omni","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","pdf"],"output":["text"]},"limit":{"context":262144,"output":64000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2024-12","release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "mimo-v2-pro-free": {"name":"MiMo V2 Pro Free","description":"Legacy model retained for compatibility with older integrations","family":"mimo","canonical_model_id":"xiaomi/mimo-v2-pro","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":64000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2024-12","release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "mimo-v2.5-free": {"name":"MiMo V2.5 Free","description":"MiMo omni model for text, image, video, audio, and agents","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "mimo-v2.6-flash-free": {"name":"MiMo-V2.6-Flash Free","description":"MiMo Flash model for multimodal coding agents and long-context automation","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "minimax-m2.1": {"name":"MiniMax-M2.1","description":"Legacy model retained for compatibility with older integrations","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.1},"knowledge":"2025-01","release_date":"2025-12-23","last_updated":"2025-12-23","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "minimax-m2.1-free": {"name":"MiniMax-M2.1 Free","description":"Legacy model retained for compatibility with older integrations","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.1","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-01","release_date":"2025-12-23","last_updated":"2025-12-23","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "minimax-m2.5": {"name":"MiniMax-M2.5","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"knowledge":"2025-01","release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "minimax-m2.5-free": {"name":"MiniMax-M2.5 Free","description":"Legacy model retained for compatibility with older integrations","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-01","release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "minimax-m2.7": {"name":"MiniMax-M2.7","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.7","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"knowledge":"2025-01","release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "minimax-m3": {"name":"MiniMax-M3","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","canonical_model_id":"minimax/MiniMax-M3","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":512000,"output":128000},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-06-01","last_updated":"2026-06-01","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "minimax-m3-free": {"name":"MiniMax-M3 Free","description":"Legacy model retained for compatibility with older integrations","family":"minimax","canonical_model_id":"minimax/MiniMax-M3","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-01","release_date":"2026-06-01","last_updated":"2026-06-01","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "minimax-m3.1-flash-free": {"name":"MiniMax M3.1 Flash Free","description":"Frontier multimodal coding model with 1M context window and tunable thinking depth","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":512000},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "mistral-large-4": {"name":"Mistral Large 4","description":"Mistral's 1T-parameter multimodal MoE for reasoning, coding, and agentic work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-4","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":524288,"output":262144},"cost":{"input":0.68,"output":2.09,"cache_read":0.07},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false,"provider":{"npm":"@ai-sdk/mistral","api":"https://opencode.ai/inference/openai/v1"}},
  "muse-spark-1.2": {"name":"Muse Spark 1.2","description":"Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows.","family":"muse","canonical_model_id":"meta/muse-spark-1.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"release_date":"2026-08-05","last_updated":"2026-08-05","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "muse-spark-1.2-contributor-free": {"name":"Muse Spark 1.2 Free","description":"Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows.","family":"muse","canonical_model_id":"meta/muse-spark-1.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-08-05","last_updated":"2026-08-05","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "muse-spark-1.3": {"name":"Muse Spark 1.3","description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for long-running agentic, multi-agent, and coding workflows. It improves long-horizon agent collaboration, instruction following, and coding efficiency relative to Muse Spark 1.2.","family":"muse","canonical_model_id":"meta/muse-spark-1.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "muse-spark-1.3-contributor-free": {"name":"Muse Spark 1.3 Free","description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for coding and agentic workflows.","family":"muse","canonical_model_id":"meta/muse-spark-1.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "muse-wonder-1.0": {"name":"Muse Wonder","description":"Muse Wonder is a frontier multimodal reasoning model from Meta for personal and professional agents, long-horizon coding, and computer use. It combines Active Memory, which carries relevant context across conversations, with Active Focus, which keeps it working through blockers toward a goal.","family":"muse","canonical_model_id":"meta/muse-wonder-1.0","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":4,"output":20,"cache_read":0.2},"release_date":"2026-10-12","last_updated":"2026-10-12","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/openai/v1"}},
  "nemotron-3-super-free": {"name":"Nemotron 3 Super Free","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-super-120b-a12b","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":128000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-06","release_date":"2026-03-11","last_updated":"2026-03-11","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "nemotron-3-ultra-free": {"name":"Nemotron 3 Ultra Free","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-ultra-550b-a55b","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-09","release_date":"2026-06-04","last_updated":"2026-06-04","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "nemotron-3.5-lightning-free": {"name":"Nemotron 3.5 Lightning Free","description":"Fast NVIDIA Nemotron MoE for reliable agentic tasks across enterprise workloads","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-lightning","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-09","release_date":"2026-08-11","last_updated":"2026-08-11","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "nemotron-3.5-super-vl-free": {"name":"Nemotron 3.5 Super VL Free","description":"NVIDIA vision-language reasoning model for text, image, and video understanding and agentic tool use","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-super-vl-120b-a12b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh"]}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"input":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "north-mini-code-free": {"name":"North Mini Code Free","description":"Cohere coding model for practical software engineering and agentic edits","family":"north","canonical_model_id":"cohere/north-mini-code-1-0","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0},"knowledge":"2025-09-23","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "qwen3-coder": {"name":"Qwen3 Coder","description":"Legacy model retained for compatibility with older integrations","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-480b-a35b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.45,"output":1.8},"knowledge":"2025-04","release_date":"2025-07-22","last_updated":"2025-07-22","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "qwen3.5-plus": {"name":"Qwen3.5 Plus","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.5-plus","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25},"knowledge":"2025-04","release_date":"2026-02-16","last_updated":"2026-02-16","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "qwen3.6-plus": {"name":"Qwen3.6 Plus","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.6-plus","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"cache_write":0.625},"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "qwen3.6-plus-free": {"name":"Qwen3.6 Plus Free","description":"Legacy model retained for compatibility with older integrations","family":"qwen","canonical_model_id":"alibaba/qwen3.6-plus","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "qwen3.8-flash": {"name":"Qwen3.8 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.47,"cache_read":0.016,"cache_write":0.2},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/anthropic/v1"}},
  "qwen3.8-max": {"name":"Qwen3.8 Max","description":"2.4-trillion-parameter MoE flagship for coding, professional work, multimodal understanding, and long-horizon agentic workflows","family":"qwen","canonical_model_id":"alibaba/qwen3.8-max","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens","min":1024}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25,"cache_write":2.5},"release_date":"2026-08-03","last_updated":"2026-08-03","open_weights":false,"provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "ring-2.6-1t-free": {"name":"Ring 2.6 1T Free","description":"Legacy model retained for compatibility with older integrations","family":"ring-1t-free","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262000,"output":66000},"cost":{"input":0,"output":0},"knowledge":"2025-06","release_date":"2026-05-08","last_updated":"2026-05-08","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "space-bunny-free": {"name":"Space Bunny Free","description":"Anonymous preview reasoning model for coding, agentic tasks, tool use, and multimodal input","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"input":524288,"output":524288},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "step-5-preview-free": {"name":"Step 5 Preview Free","description":"StepFun's next-generation flagship base model for coding and professional knowledge work, with native text, image, and video input and a 1M-token context window","canonical_model_id":"stepfun/step-5-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-09-16","last_updated":"2026-09-20","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "trinity-large-preview-free": {"name":"Trinity Large Preview","description":"Legacy model retained for compatibility with older integrations","family":"trinity","canonical_model_id":"arcee-ai/trinity-large-preview","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0,"output":0},"knowledge":"2025-06","release_date":"2026-01-27","last_updated":"2026-05-28","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}},
  "x-preview-f-free": {"name":"Ox Alpha Free (Unlimited)","description":"Stealth reasoning model for coding, agentic tasks, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-08-21","last_updated":"2026-08-21","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/openai/v1"}}
}},
"opencode-go": {"models": {
  "beam-free": {"name":"Beam Free","description":"Sparse mixture-of-experts reasoning model with 23B active parameters for coding and agentic workflows","family":"beam","canonical_model_id":"reflection/beam-1.0-501b-a23b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"knowledge":"2026-06-30","release_date":"2026-10-05","last_updated":"2026-10-16","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "cat-jam-free": {"name":"Cat Jam Free","description":"A text-only reasoning model for coding and agentic workflows, with tool calling and structured JSON outputs.","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-10-12","last_updated":"2026-10-12","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "claude-haiku-5-5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/go/anthropic/v1"}},
  "deepseek-v4-pro": {"name":"DeepSeek V4 Pro (New)","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":0.66,"output":1.98,"cache_read":0.022},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "deepseek-v4.1-flash": {"name":"DeepSeek V4.1 Flash","description":"DeepSeek V4.1 Flash model for reasoning and agentic coding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":384000},"cost":{"input":0.15,"output":0.6,"cache_read":0.003},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "glm-5.2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "glm-5.3": {"name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "glm-5.3-flash": {"name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"Cost-efficient GPT-5.6 model for fast, high-volume workloads","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/go/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "gpt-6-luna": {"name":"GPT-6 Luna","description":"OpenAI's most efficient model for focused, high-volume tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/go/openai/v1"},"rig":{"reasoning_default":"medium","sampling":"reasoning_off"}},
  "grok-4.7": {"name":"Grok 4.7","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/go/openai/v1"}},
  "hy3": {"name":"Hy3","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"input":192000,"output":128000},"cost":{"input":0.14,"output":0.58,"cache_read":0.035},"release_date":"2026-07-06","last_updated":"2026-07-06","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "hy4-preview": {"name":"Hy4 preview","description":"A next-generation productivity model with significantly enhanced Agent and complex task execution capabilities.","family":"Hy","canonical_model_id":"tencent/hy4-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high"]}],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1024000,"output":64000},"cost":{"input":0.834,"output":2.501,"cache_read":0.042},"release_date":"2026-08-28","last_updated":"2026-08-28","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "iceman-contributor-free": {"name":"Iceman Contributor Free","description":"Stealth model, free for a limited time","reasoning":true,"reasoning_options":[],"tool_call":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-10-24","last_updated":"2026-10-24","open_weights":false,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "kimi-k2.7-code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.95,"output":4,"cache_read":0.19},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "kimi-k3": {"name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "kimi-k3.1-pro": {"name":"Kimi K3.1 Pro","description":"Multimodal Kimi model with 1M context and always-on reasoning for coding, 3D creation, financial research, and agentic work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3.1-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":false,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "longcat-2.0": {"name":"LongCat-2.0","description":"Meituan LongCat-2.0, a reasoning model with tool calling and a 1M-token context window","family":"longcat","canonical_model_id":"meituan/longcat-2.0","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.006},"release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "longcat-2.5-preview-free": {"name":"LongCat 2.5 Preview Free","description":"Meituan's multimodal reasoning model for coding and agent tasks, with image understanding and a 1M-token context window","family":"longcat","canonical_model_id":"meituan/longcat-2.5-preview","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-09-25","last_updated":"2026-09-25","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "mimo-v2.5": {"name":"MiMo V2.5","description":"MiMo omni model for text, image, video, audio, and agents","family":"mimo-v2.5","canonical_model_id":"xiaomi/mimo-v2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "mimo-v2.5-pro": {"name":"MiMo V2.5 Pro","description":"MiMo pro model for strong multimodal reasoning and agent execution","family":"mimo-v2.5-pro","canonical_model_id":"xiaomi/mimo-v2.5-pro","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":128000},"cost":{"input":0.435,"output":0.87,"cache_read":0.003625},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "mimo-v2.6-flash": {"name":"MiMo-V2.6-Flash","description":"MiMo Flash model for multimodal coding agents and long-context automation","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "mimo-v2.6-pro": {"name":"MiMo-V2.6-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-pro","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.435,"output":0.87,"cache_read":0.003625},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "minimax-m2.7": {"name":"MiniMax-M2.7","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.7","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06,"cache_write":0.375},"knowledge":"2025-01","release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/go/anthropic/v1"}},
  "minimax-m3": {"name":"MiniMax-M3","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax-m3","canonical_model_id":"minimax/MiniMax-M3","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06,"tiers":[{"input":0.6,"output":2.4,"cache_read":0.12,"tier":{"type":"context","size":512000}}],"context_over_200k":{"input":0.6,"output":2.4,"cache_read":0.12}},"knowledge":"2025-01","release_date":"2026-06-01","last_updated":"2026-06-01","open_weights":true,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/go/anthropic/v1"}},
  "minimax-m3.1-flash": {"name":"MiniMax M3.1 Flash","description":"Frontier multimodal coding model with 1M context window and tunable thinking depth","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":512000},"cost":{"input":0.15,"output":0.6,"cache_read":0.015},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "minimax-m3.1-flash-free": {"name":"MiniMax M3.1 Flash Free","description":"Frontier multimodal coding model with 1M context window and tunable thinking depth","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":512000},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "muse-spark-1.2-contributor": {"name":"Muse Spark 1.2 Contributor","description":"Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows.","family":"muse","canonical_model_id":"meta/muse-spark-1.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.1,"output":0.2,"cache_read":0.002},"release_date":"2026-08-05","last_updated":"2026-08-05","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/go/openai/v1"}},
  "muse-spark-1.3-contributor": {"name":"Muse Spark 1.3 Contributor","description":"Muse Spark 1.3 is a multimodal reasoning model from Meta for coding and agentic workflows.","family":"muse","canonical_model_id":"meta/muse-spark-1.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.1,"output":0.2,"cache_read":0.002},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"provider":{"npm":"@ai-sdk/openai","api":"https://opencode.ai/inference/go/openai/v1"}},
  "qwen3.7-plus": {"name":"Qwen3.7 Plus","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"qwen3.7-plus","canonical_model_id":"alibaba/qwen3.7-plus","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","max":262144}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.4,"output":1.6,"cache_read":0.04,"cache_write":0.5,"tiers":[{"input":1.2,"output":4.8,"cache_read":0.12,"cache_write":1.5,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":1.2,"output":4.8,"cache_read":0.12,"cache_write":1.5}},"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/go/anthropic/v1"}},
  "qwen3.8-flash": {"name":"Qwen3.8 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.47,"cache_read":0.016,"cache_write":0.2},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/go/anthropic/v1"}},
  "qwen3.8-max": {"name":"Qwen3.8 Max","description":"2.4-trillion-parameter multimodal flagship for coding, professional work, and long-horizon agentic workflows","family":"qwen3.8-max","canonical_model_id":"alibaba/qwen3.8-max","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","xhigh"]},{"type":"budget_tokens","max":262144}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25,"cache_write":2.5},"release_date":"2026-08-03","last_updated":"2026-08-03","open_weights":false,"provider":{"npm":"@ai-sdk/anthropic","api":"https://opencode.ai/inference/go/anthropic/v1"}},
  "space-bunny": {"name":"Space Bunny","description":"Anonymous preview reasoning model for coding, agentic tasks, tool use, and multimodal input","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"input":524288,"output":524288},"cost":{"input":0.15,"output":0.6,"cache_read":0.03,"cache_write":0},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "step-5-preview": {"name":"Step 5 Preview","description":"StepFun's next-generation flagship base model for coding and professional knowledge work, with native text, image, and video input and a 1M-token context window","canonical_model_id":"stepfun/step-5-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":1,"output":2.7,"cache_read":0.1},"release_date":"2026-09-16","last_updated":"2026-09-20","open_weights":true,"provider":{"api":"https://opencode.ai/inference/go/openai/v1"}},
  "step-5-preview-free": {"name":"Step 5 Preview Free","description":"StepFun's next-generation flagship base model for coding and professional knowledge work, with native text, image, and video input and a 1M-token context window","canonical_model_id":"stepfun/step-5-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"input":1000000,"output":65536},"cost":{"input":0,"output":0,"cache_read":0},"release_date":"2026-09-16","last_updated":"2026-09-20","open_weights":true,"status":"deprecated","provider":{"api":"https://opencode.ai/inference/go/openai/v1"}}
}},
"openrouter": {"models": {
  "aion-labs/aion-2.0": {"name":"Aion-2.0","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","canonical_model_id":"aion-labs/aion-2.0","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.8,"output":1.6,"cache_read":0.2},"release_date":"2025-12-21","last_updated":"2025-12-21","open_weights":false},
  "aion-labs/aion-3.0": {"name":"Aion-3.0","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","canonical_model_id":"aion-labs/aion-3.0","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":3,"output":6,"cache_read":0.75},"release_date":"2026-05-05","last_updated":"2026-05-05","open_weights":false},
  "aion-labs/aion-3.0-mini": {"name":"Aion-3.0-Mini","description":"Efficient model for low-latency assistance, extraction, and routine automation","canonical_model_id":"aion-labs/aion-3.0-mini","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.7,"output":1.4,"cache_read":0.18},"release_date":"2026-05-14","last_updated":"2026-05-14","open_weights":false},
  "aion-labs/aion-3.5": {"name":"Aion 3.5","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":3,"output":6,"cache_read":0.75},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "aion-labs/aion-3.5-mini": {"name":"Aion 3.5 Mini","description":"Efficient model for low-latency assistance, extraction, and routine automation","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.7,"output":1.4,"cache_read":0.18},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "aion-labs/aion-rp-llama-3.1-8b": {"name":"Aion-RP 1.0 (8B)","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"aion-labs/aion-rp-llama-3.1-8b","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":29491},"cost":{"input":0.8,"output":1.6},"knowledge":"2023-12-31","release_date":"2025-02-04","last_updated":"2025-02-04","open_weights":true},
  "amazon/nova-2-lite-v1": {"name":"Nova 2 Lite","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"nova","canonical_model_id":"amazon/nova-2-lite-v1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":65535},"cost":{"input":0.3,"output":2.5},"knowledge":"2025-10","release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "amazon/nova-lite-v1": {"name":"Nova Lite 1.0","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-lite","canonical_model_id":"amazon/nova-lite-v1","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":300000,"output":5120},"cost":{"input":0.06,"output":0.24},"knowledge":"2024-10-31","release_date":"2024-12-05","last_updated":"2024-12-05","open_weights":false},
  "amazon/nova-micro-v1": {"name":"Nova Micro 1.0","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"nova-micro","canonical_model_id":"amazon/nova-micro-v1","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":5120},"cost":{"input":0.035,"output":0.14},"knowledge":"2024-10-31","release_date":"2024-12-05","last_updated":"2024-12-05","open_weights":false},
  "amazon/nova-premier-v1": {"name":"Nova Premier 1.0","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova","canonical_model_id":"amazon/nova-premier","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":32000},"cost":{"input":2.5,"output":12.5,"cache_read":0.625},"knowledge":"2024-10","release_date":"2025-04-30","last_updated":"2025-04-30","open_weights":false,"status":"deprecated"},
  "amazon/nova-pro-v1": {"name":"Nova Pro 1.0","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"nova-pro","canonical_model_id":"amazon/nova-pro-v1","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":300000,"output":5120},"cost":{"input":0.8,"output":3.2},"knowledge":"2024-10-31","release_date":"2024-12-05","last_updated":"2024-12-05","open_weights":false},
  "anthracite-org/magnum-v4-72b": {"name":"Magnum v4 72B","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","canonical_model_id":"anthracite-org/magnum-v4-72b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":4096},"cost":{"input":2.5,"output":5},"knowledge":"2024-06-30","release_date":"2024-10-22","last_updated":"2024-10-22","open_weights":true},
  "anthropic/claude-3.7-sonnet": {"name":"Claude 3.7 Sonnet","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "anthropic/claude-fable-5": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic/claude-fable-5.1": {"name":"Claude Fable 5.1","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never"}},
  "anthropic/claude-haiku-4.5": {"name":"Claude Haiku 4.5 (latest)","description":"Fast Claude lane for lightweight agents, office tasks, and responsive chat","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic/claude-haiku-5.5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "anthropic/claude-opus-4.1": {"name":"Claude Opus 4.1 (latest)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic/claude-opus-4.5": {"name":"Claude Opus 4.5 (latest)","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic/claude-opus-4.6": {"name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "anthropic/claude-opus-4.7": {"name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25,"tiers":[{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":10,"output":37.5,"cache_read":1,"cache_write":12.5}},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic/claude-opus-4.8": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic/claude-opus-5": {"name":"Claude Opus 5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic/claude-opus-5.5": {"name":"Claude Opus 5.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "anthropic/claude-sonnet-4.5": {"name":"Claude Sonnet 4.5 (latest)","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "anthropic/claude-sonnet-4.6": {"name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75,"tiers":[{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":6,"output":22.5,"cache_read":0.6,"cache_write":7.5}},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "anthropic/claude-sonnet-5": {"name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "anthropic/claude-sonnet-5.5": {"name":"Claude Sonnet 5.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "bytedance-seed/seed-1.6": {"name":"Seed 1.6","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","canonical_model_id":"bytedance-seed/seed-1-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.25,"output":2,"tiers":[{"input":0.5,"output":4,"tier":{"type":"context","size":128000}}]},"release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "bytedance-seed/seed-1.6-flash": {"name":"Seed 1.6 Flash","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","canonical_model_id":"bytedance-seed/seed-1-6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.075,"output":0.3,"tiers":[{"input":0.1,"output":0.8,"tier":{"type":"context","size":128000}}]},"release_date":"2025-08-28","last_updated":"2025-08-28","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "bytedance-seed/seed-2-1-turbo": {"name":"Seed 2.1 Turbo","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","canonical_model_id":"bytedance-seed/seed-2.1-turbo","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.5,"output":2.5},"release_date":"2026-06-23","last_updated":"2026-06-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "bytedance-seed/seed-2.0-code": {"name":"Seed 2.0 Code","description":"Coding model for repository understanding, refactors, and agentic engineering tasks","family":"seed","canonical_model_id":"bytedance-seed/seed-2.0-code","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.5,"output":3,"tiers":[{"input":1,"output":6,"tier":{"type":"context","size":128000}}]},"release_date":"2026-02-14","last_updated":"2026-02-14","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "bytedance-seed/seed-2.0-lite": {"name":"Seed 2.0 Lite","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","canonical_model_id":"bytedance-seed/seed-2.0-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.25,"output":2,"tiers":[{"input":0.5,"output":4,"tier":{"type":"context","size":128000}}]},"release_date":"2026-02-14","last_updated":"2026-02-14","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "bytedance-seed/seed-2.0-mini": {"name":"Seed 2.0 Mini","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","canonical_model_id":"bytedance-seed/seed-2.0-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.1,"output":0.4,"tiers":[{"input":0.2,"output":0.8,"tier":{"type":"context","size":128000}}]},"release_date":"2026-02-14","last_updated":"2026-02-14","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "bytedance/ui-tars-1.5-7b": {"name":"UI-TARS 7B ","description":"Multimodal model for analyzing text, images, documents, and rich media","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text"],"output":["text"]},"limit":{"context":128000,"output":2048},"cost":{"input":0.1,"output":0.2,"cache_read":0.1},"knowledge":"2025-01-31","release_date":"2025-07-22","last_updated":"2025-07-22","open_weights":true},
  "cognitivecomputations/dolphin-mistral-24b-venice-edition": {"name":"Uncensored","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral","canonical_model_id":"cognitivecomputations/dolphin-mistral-24b-venice-edition","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.2,"output":0.9},"knowledge":"2024-04-30","release_date":"2025-07-09","last_updated":"2025-07-09","open_weights":true},
  "cohere/command-a": {"name":"Command A","description":"Cohere command model for multilingual enterprise agents, tools, and chat","family":"command-a","canonical_model_id":"cohere/command-a","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":8192},"cost":{"input":2.5,"output":10},"knowledge":"2024-08-31","release_date":"2025-03-13","last_updated":"2025-03-13","open_weights":true},
  "cohere/command-a-plus": {"name":"Command A+","description":"Cohere command model for multilingual enterprise agents, tools, and chat","family":"command-a","canonical_model_id":"cohere/command-a-plus-05-2026","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":192000,"output":64000},"cost":{"input":0.3,"output":1.5,"cache_read":0.15},"knowledge":"2025-04-01","release_date":"2026-05-20","last_updated":"2026-06-09","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "cohere/command-r-08-2024": {"name":"Command R","description":"Cohere retrieval model for long-context chat and enterprise RAG workflows","family":"command-r","canonical_model_id":"cohere/command-r-08-2024","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":0.15,"output":0.6},"knowledge":"2024-06-01","release_date":"2024-08-30","last_updated":"2024-08-30","open_weights":true},
  "cohere/command-r-plus-08-2024": {"name":"Command R+","description":"Cohere's RAG workhorse for long-context enterprise search and tool use","family":"command-r","canonical_model_id":"cohere/command-r-plus-08-2024","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":2.5,"output":10},"knowledge":"2024-06-01","release_date":"2024-08-30","last_updated":"2024-08-30","open_weights":true},
  "cohere/command-r7b-12-2024": {"name":"Command R7B","description":"Cohere retrieval model for long-context chat and enterprise RAG workflows","family":"command-r","canonical_model_id":"cohere/command-r7b-12-2024","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4000},"cost":{"input":0.0375,"output":0.15},"knowledge":"2024-06-01","release_date":"2024-12-13","last_updated":"2024-12-13","open_weights":true},
  "cohere/north-mini-code:free": {"name":"North Mini Code (free)","description":"Cohere coding model for practical software engineering and agentic edits","family":"north","canonical_model_id":"cohere/north-mini-code-1-0","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":64000},"cost":{"input":0,"output":0},"knowledge":"2025-09-23","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-chat": {"name":"DeepSeek Chat","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-chat","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":16000},"cost":{"input":0.2574,"output":1.0287},"knowledge":"2025-09","release_date":"2025-12-01","last_updated":"2026-02-28","open_weights":true},
  "deepseek/deepseek-chat-v3-0324": {"name":"DeepSeek V3 0324","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-chat-v3-0324","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":115200},"cost":{"input":0.29,"output":1.14,"cache_read":0.11},"knowledge":"2024-07-31","release_date":"2025-03-24","last_updated":"2025-03-24","open_weights":true},
  "deepseek/deepseek-chat-v3.1": {"name":"DeepSeek V3.1","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-chat-v3.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":147456},"cost":{"input":0.25,"output":0.95,"cache_read":0.13},"knowledge":"2025-03-31","release_date":"2025-08-21","last_updated":"2025-08-21","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-r1-0528": {"name":"DeepSeek R1 0528","description":"DeepSeek reasoning model for multi-step analysis, math, coding, and tools","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-r1-0528","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":32768},"cost":{"input":0.5,"output":2.15,"cache_read":0.35},"release_date":"2025-05-28","last_updated":"2025-05-28","open_weights":true},
  "deepseek/deepseek-v3.1-terminus": {"name":"DeepSeek V3.1 Terminus","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":147456},"cost":{"input":0.27,"output":1},"knowledge":"2025-03-31","release_date":"2025-09-22","last_updated":"2025-09-22","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v3.2": {"name":"DeepSeek V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":147456},"cost":{"input":0.26,"output":0.38,"cache_read":0.13},"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v3.2-exp": {"name":"DeepSeek V3.2 Exp","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":147456},"cost":{"input":0.27,"output":0.41},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v4-flash": {"name":"DeepSeek V4 Flash","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":943717},"cost":{"input":0.091,"output":0.182,"cache_read":0.0182},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v4-flash-0731": {"name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-0731","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.119,"output":0.238,"cache_read":0.0238},"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v4-flash-vision-exp": {"name":"DeepSeek V4 Flash Vision Exp","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-vision-exp","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0.2156,"output":0.6468,"cache_read":0.00686},"release_date":"2026-08-21","last_updated":"2026-09-01","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v4-pro": {"name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":1.6,"output":3.2,"cache_read":0.135},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v4-pro-0813": {"name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro-0813","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.66,"output":1.98,"cache_read":0.022},"release_date":"2026-08-13","last_updated":"2026-08-22","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v4.1-flash": {"name":"DeepSeek V4.1 Flash","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.045,"output":0.5,"cache_read":0.006},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "deepseek/deepseek-v4.1-flash-1017": {"name":"DeepSeek V4.1 Flash 1017","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.15,"output":0.6,"cache_read":0.003},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "dots-studio/dots-3-note-preview:free": {"name":"Dots3-Note Preview (free)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":512000,"output":460800},"cost":{"input":0,"output":0},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "elorian/glimpse": {"name":"Glimpse","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":256000},"cost":{"input":0.6,"output":3,"cache_read":0.06},"release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "fireworks/ember-1": {"name":"Ember-1","description":"Multimodal reasoning model for visual analysis, planning, and tool use","canonical_model_id":"fireworks/ember-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":3,"output":15,"cache_read":0.3},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/chirp-3": {"name":"Chirp 3","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "google/gemini-2.0-flash-001": {"name":"Gemini 2.0 Flash","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "google/gemini-2.5-flash": {"name":"Gemini 2.5 Flash","description":"Fast Gemini workhorse for multimodal apps where latency and price matter","family":"gemini-flash","canonical_model_id":"google/gemini-2.5-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65535},"cost":{"input":0.3,"output":2.5,"reasoning":2.5,"cache_read":0.03,"cache_write":0.083333,"input_audio":1},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "google/gemini-2.5-flash-image": {"name":"Nano Banana","description":"Nano Banana image model for fast generation, edits, and character-consistent assets","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-2.5-flash-image","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":32768,"output":8192},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"cache_write":0.083333,"input_audio":1},"knowledge":"2024-06","release_date":"2025-08-26","last_updated":"2025-08-26","open_weights":false},
  "google/gemini-2.5-flash-lite": {"name":"Gemini 2.5 Flash-Lite","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","canonical_model_id":"google/gemini-2.5-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65535},"cost":{"input":0.1,"output":0.4,"reasoning":0.4,"cache_read":0.01,"cache_write":0.083333,"input_audio":0.3},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "google/gemini-2.5-pro": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"reasoning":10,"cache_read":0.125,"cache_write":0.375,"input_audio":1.25,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "google/gemini-2.5-pro-preview": {"name":"Gemini 2.5 Pro Preview 06-05","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini","canonical_model_id":"google/gemini-2.5-pro-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["pdf","image","text","audio"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"reasoning":10,"cache_read":0.125,"cache_write":0.375,"input_audio":1.25,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01-31","release_date":"2025-06-05","last_updated":"2025-06-05","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3-flash-preview": {"name":"Gemini 3 Flash Preview","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","canonical_model_id":"google/gemini-3-flash-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.5,"output":3,"reasoning":3,"cache_read":0.05,"cache_write":0.083333,"input_audio":1},"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3-pro-image": {"name":"Nano Banana Pro","description":"Nano Banana Pro for higher-fidelity image generation and design-heavy edits","family":"gemini-pro","type":"image","canonical_model_id":"google/gemini-3-pro-image","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":131072,"output":32768},"cost":{"input":2,"output":12,"reasoning":12,"cache_read":0.2,"cache_write":0.375,"input_audio":2},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "google/gemini-3-pro-image-preview": {"name":"Nano Banana Pro Preview","description":"Nano Banana Pro for higher-fidelity image generation and design-heavy edits","family":"gemini-pro","type":"image","canonical_model_id":"google/gemini-3-pro-image-preview","reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":65536,"output":32768},"cost":{"input":2,"output":12,"reasoning":12,"cache_read":0.2,"cache_write":0.375,"input_audio":2},"knowledge":"2025-01","release_date":"2025-11-20","last_updated":"2025-11-20","open_weights":false},
  "google/gemini-3.1-flash-image": {"name":"Nano Banana 2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-3.1-flash-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text"],"output":["text","image"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.5,"output":3},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.1-flash-image-preview": {"name":"Nano Banana 2 Preview","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-3.1-flash-image-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text"],"output":["text","image"]},"limit":{"context":65536,"output":58982},"cost":{"input":0.5,"output":3},"knowledge":"2025-01","release_date":"2026-02-26","last_updated":"2026-02-26","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.1-flash-lite": {"name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.1-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"reasoning":1.5,"cache_read":0.025,"cache_write":0.083333,"input_audio":0.5},"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.1-flash-lite-image": {"name":"Nano Banana 2 Lite","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gemini-flash-lite","type":"image","canonical_model_id":"google/gemini-3.1-flash-lite-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","image"]},"limit":{"context":65536,"output":58982},"cost":{"input":0.25,"output":1.5},"knowledge":"2025-01","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.1-flash-lite-preview": {"name":"Gemini 3.1 Flash Lite Preview","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.1-flash-lite-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"reasoning":1.5,"cache_read":0.025,"cache_write":0.083333,"input_audio":0.5},"knowledge":"2025-01","release_date":"2026-03-03","last_updated":"2026-03-03","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.1-pro-preview": {"name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","canonical_model_id":"google/gemini-3.1-pro-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"reasoning":12,"cache_read":0.2,"cache_write":0.375,"input_audio":2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.1-pro-preview-customtools": {"name":"Gemini 3.1 Pro Preview Custom Tools","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-3.1-pro-preview-customtools","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"reasoning":12,"cache_read":0.2,"cache_write":0.375,"input_audio":2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.5-flash": {"name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.5-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"reasoning":9,"cache_read":0.15,"cache_write":0.083333,"input_audio":3},"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.5-flash-lite": {"name":"Gemini 3.5 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.5-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"reasoning":2.5,"cache_read":0.03,"cache_write":0.083333,"input_audio":0.3},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.6-flash": {"name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"reasoning":3.75,"cache_read":0.075,"cache_write":0.041667,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.7-flash": {"name":"Gemini 3.7 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"reasoning":3.75,"cache_read":0.075,"cache_write":0.041667,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-3.8-flash": {"name":"Gemini 3.8 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.8-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"reasoning":3.75,"cache_read":0.075,"cache_write":0.041667,"input_audio":0.75},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-4-argon": {"name":"Gemini 4 Argon","description":"Gemini multimodal model for text, image, audio, video, and document tasks","family":"gemini","canonical_model_id":"google/gemini-4-argon","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":2,"output":10,"reasoning":10,"cache_read":0.1,"cache_write":0.1875,"input_audio":2},"knowledge":"2026-03","release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemini-nano-banana-2.1": {"name":"Nano Banana 2.1","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-nano-banana-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text"],"output":["text","image"]},"limit":{"context":65536,"output":58982},"cost":{"input":1.5,"output":7.5},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "google/gemma-2-27b-it": {"name":"Gemma 2 27B","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-2-27b-it","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":2048},"cost":{"input":0.65,"output":0.65},"knowledge":"2024-06-30","release_date":"2024-07-13","last_updated":"2024-07-13","open_weights":true},
  "google/gemma-3-12b-it": {"name":"Gemma 3 12B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-3-12b-it","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.05,"output":0.15},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true},
  "google/gemma-3-27b-it": {"name":"Gemma 3 27B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-3-27b-it","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":0.08,"output":0.35,"cache_read":0.04},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true},
  "google/gemma-3-4b-it": {"name":"Gemma 3 4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-3-4b-it","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.05,"output":0.1},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true},
  "google/gemma-4-26b-a4b-it": {"name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-26b-a4b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text","video"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.0675,"output":0.225,"cache_read":0.0375},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "google/gemma-4-26b-a4b-it:free": {"name":"Gemma 4 26B A4B  (free)","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-26b-a4b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["image","text","video"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "google/gemma-4-31b-it": {"name":"Gemma 4 31B IT","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","canonical_model_id":"google/gemma-4-31b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text","video"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":0.2,"output":0.4},"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "google/gemma-4-31b-it:free": {"name":"Gemma 4 31B (free)","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","canonical_model_id":"google/gemma-4-31b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["image","text","video"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0},"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "google/lyria-3-clip-preview": {"name":"Lyria 3 Clip Preview","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"lyria","canonical_model_id":"google/lyria-3-clip-preview","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","audio"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0,"output":0},"release_date":"2026-03-25","last_updated":"2026-03-25","open_weights":false},
  "google/lyria-3-pro-preview": {"name":"Lyria 3 Pro Preview","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"lyria","canonical_model_id":"google/lyria-3-pro-preview","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text","audio"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0,"output":0},"release_date":"2026-03-25","last_updated":"2026-03-25","open_weights":false},
  "gryphe/mythomax-l2-13b": {"name":"MythoMax L2 13B","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","canonical_model_id":"gryphe/mythomax-l2-13b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":3686},"cost":{"input":0.08,"output":0.11},"release_date":"2023-08-10","last_updated":"2023-08-10","open_weights":true},
  "hexgrad/kokoro-82m": {"name":"Kokoro 82M","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "ibm-granite/granite-4.2-8b": {"name":"Granite 4.2 8B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"granite","canonical_model_id":"ibm/granite-4.2-8b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":0.06,"output":0.15,"cache_read":0.015},"release_date":"2026-08-25","last_updated":"2026-08-25","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "inception/mercury-2": {"name":"Mercury 2","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"mercury","canonical_model_id":"inception/mercury-2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":50000},"cost":{"input":0.25,"output":0.75,"cache_read":0.025},"release_date":"2026-02-24","last_updated":"2026-02-24","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "inception/mercury-2.5": {"name":"Mercury 2.5","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"mercury","canonical_model_id":"inception/mercury-2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":260000,"output":65536},"cost":{"input":0.04,"output":0.15,"cache_read":0.004},"release_date":"2026-09-08","last_updated":"2026-09-08","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "inception/mercury-3": {"name":"Mercury 3","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"mercury","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.1,"output":0.35,"cache_read":0.01},"release_date":"2026-10-16","last_updated":"2026-10-16","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "inclusionai/ling-3.0-flash": {"name":"Ling 3.0 Flash","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","canonical_model_id":"inclusionai/ling-3.0-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.021,"output":0.063,"cache_read":0.0042},"release_date":"2026-07-23","last_updated":"2026-07-23","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "inclusionai/ling-3.0-flash-vl": {"name":"Ling 3.0 Flash VL","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"ling","canonical_model_id":"inclusionai/ling-3.0-flash-vl","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.06,"output":0.18,"cache_read":0.012},"release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "inclusionai/ling-3.1-flash": {"name":"Ling 3.1 Flash","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"ling","canonical_model_id":"inclusionai/ling-3.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":262144},"cost":{"input":0.075,"output":0.225,"cache_read":0.015},"release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "inference-net/schematron-v2-small": {"name":"Schematron V2 Small","description":"Efficient model for low-latency assistance, extraction, and routine automation","canonical_model_id":"inference-net/schematron-v2-small","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.05,"output":0.23,"cache_read":0.05},"release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false},
  "inference-net/schematron-v2-turbo": {"name":"Schematron V2 Turbo","description":"Efficient model for low-latency assistance, extraction, and routine automation","canonical_model_id":"inference-net/schematron-v2-turbo","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.03,"output":0.15,"cache_read":0.03},"release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false},
  "liquid/lfm-2.5-2.6b": {"name":"LFM2.5-2.6B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"liquid","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.35,"output":0.5,"cache_read":0.04},"release_date":"2026-08-11","last_updated":"2026-08-11","open_weights":true},
  "liquid/lfm-2.5-2.6b:free": {"name":"LFM2.5-2.6B (free)","description":"Free provider route for experiments, demos, and cost-sensitive chat workloads","family":"liquid","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":65536,"output":8192},"cost":{"input":0,"output":0},"release_date":"2026-08-11","last_updated":"2026-08-11","open_weights":true},
  "liquid/lfm-2.5-24b-a2b": {"name":"LFM2.5-24B-A2B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"liquid","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.05,"output":0.7,"cache_read":0.025},"release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":true},
  "mancer/weaver": {"name":"Weaver (alpha)","description":"General-purpose chat model for instruction following, writing, and analysis","family":"alpha","canonical_model_id":"mancer/weaver","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8000,"output":6000},"cost":{"input":0.4,"output":0.75},"knowledge":"2023-06-30","release_date":"2023-08-02","last_updated":"2023-08-02","open_weights":false},
  "meituan/longcat-2.0": {"name":"LongCat-2.0","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"longcat","canonical_model_id":"meituan/longcat-2.0","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048756,"output":262144},"cost":{"input":0.3,"output":1.2,"cache_read":0.006},"release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "meta-llama/llama-3.1-70b-instruct": {"name":"Llama-3.1-70B-Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.1-70b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.4,"output":0.4},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true},
  "meta-llama/llama-3.1-8b-instruct": {"name":"Llama-3.1-8B-Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.1-8b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":0.05,"output":0.08,"cache_read":0.025},"knowledge":"2023-12","release_date":"2024-07-23","last_updated":"2024-07-23","open_weights":true},
  "meta-llama/llama-3.2-1b-instruct": {"name":"Llama 3.2 1B Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.2-1b-instruct","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":60000,"output":54000},"cost":{"input":0.027,"output":0.201},"knowledge":"2023-12","release_date":"2024-09-25","last_updated":"2024-09-25","open_weights":true},
  "meta-llama/llama-3.2-3b-instruct": {"name":"Llama 3.2 3B Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.2-3b-instruct","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":0.05,"output":0.33},"knowledge":"2023-12","release_date":"2024-09-25","last_updated":"2024-09-25","open_weights":true},
  "meta-llama/llama-3.3-70b-instruct": {"name":"Llama-3.3-70B-Instruct","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","canonical_model_id":"meta/llama-3.3-70b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.1,"output":0.32},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true},
  "meta-llama/llama-4-maverick": {"name":"Llama 4 Maverick","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","canonical_model_id":"meta/llama-4-maverick-17b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":524288,"output":16384},"cost":{"input":0.1875,"output":0.6525,"cache_read":0.05},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true},
  "meta-llama/llama-4-scout": {"name":"Llama 4 Scout","description":"Open multimodal Llama model for long-context analysis and efficient agents","family":"llama","canonical_model_id":"meta/llama-4-scout-17b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":327680,"output":16384},"cost":{"input":0.1,"output":0.3},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true},
  "meta-llama/llama-guard-4-12b": {"name":"Llama Guard 4 12B","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"llama","canonical_model_id":"meta/llama-guard-4-12b","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":163840,"output":16384},"cost":{"input":0.18,"output":0.18},"release_date":"2025-04-29","last_updated":"2025-04-29","open_weights":true},
  "meta/muse-glimmer-30b": {"name":"Muse Glimmer 30B","description":"Muse Glimmer is a 30-billion-parameter open-weight multimodal model from Meta Superintelligence Labs, distilled from Muse Spark for always-on local agents, tool use, coding, and image understanding.","family":"muse","canonical_model_id":"meta/muse-glimmer-30b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.3,"output":1.2,"cache_read":0.04},"knowledge":"2026-01-04","release_date":"2026-08-10","last_updated":"2026-08-10","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "meta/muse-spark-1.1": {"name":"Muse Spark 1.1","description":"Open Llama multimodal model for image understanding and text reasoning","family":"muse","canonical_model_id":"meta/muse-spark-1.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf","video"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"release_date":"2026-04-08","last_updated":"2026-07-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "meta/muse-spark-1.2": {"name":"Muse Spark 1.2","description":"Muse Spark 1.2 is a coding-focused update to Muse Spark 1.1 with improvements in code generation, complex debugging, codebase understanding, and end-to-end developer workflows.","family":"muse","canonical_model_id":"meta/muse-spark-1.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"release_date":"2026-08-05","last_updated":"2026-08-05","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "meta/muse-spark-1.2-contributor": {"name":"Muse Spark 1.2 Contributor","description":"Open Llama multimodal model for image understanding and text reasoning","family":"muse","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.1,"output":0.2,"cache_read":0.002},"release_date":"2026-08-21","last_updated":"2026-08-21","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "meta/muse-spark-1.3": {"name":"Muse Spark 1.3","description":"Open Llama multimodal model for image understanding and text reasoning","family":"muse","canonical_model_id":"meta/muse-spark-1.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "meta/muse-spark-1.3-contributor": {"name":"Muse Spark 1.3 Contributor","description":"Open Llama multimodal model for image understanding and text reasoning","family":"muse","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.1,"output":0.2,"cache_read":0.002},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "meta/muse-spark-open": {"name":"Muse Spark Open","description":"Open Llama multimodal model for image understanding and text reasoning","family":"muse","canonical_model_id":"meta/muse-spark-open","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"knowledge":"2026-01-04","release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "meta/muse-wonder": {"name":"Muse Wonder","description":"Open Llama multimodal model for image understanding and text reasoning","family":"muse","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":4,"output":20,"cache_read":0.2},"release_date":"2026-10-12","last_updated":"2026-10-12","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "microsoft/phi-4": {"name":"Phi 4","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","family":"phi","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":16384,"output":14745},"cost":{"input":0.07,"output":0.14},"knowledge":"2024-06-30","release_date":"2025-01-10","last_updated":"2025-01-10","open_weights":true},
  "minimax/minimax-01": {"name":"MiniMax-01","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000192,"output":40000},"cost":{"input":0.2,"output":1.1},"knowledge":"2024-03-31","release_date":"2025-01-15","last_updated":"2025-01-15","open_weights":true},
  "minimax/minimax-m1": {"name":"MiniMax M1","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":40000},"cost":{"input":0.55,"output":2.2},"knowledge":"2024-06-30","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "minimax/minimax-m2": {"name":"MiniMax-M2","description":"Efficient open MiniMax model built for coding agents and tool-heavy workflows","family":"minimax","canonical_model_id":"minimax/MiniMax-M2","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.255,"output":1.02},"release_date":"2025-10-27","last_updated":"2025-10-27","open_weights":true},
  "minimax/minimax-m2-her": {"name":"MiniMax-M2 Her","description":"MiniMax model for chat, coding, office work, and agentic tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2-Her","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":65536,"output":2048},"cost":{"input":0.3,"output":1.2,"cache_read":0.03},"release_date":"2026-01-23","last_updated":"2026-01-23","open_weights":false},
  "minimax/minimax-m2.1": {"name":"MiniMax-M2.1","description":"Earlier MiniMax agent model for practical coding and productivity tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.1","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03},"release_date":"2025-12-23","last_updated":"2025-12-23","open_weights":true},
  "minimax/minimax-m2.5": {"name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.03},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true},
  "minimax/minimax-m2.7": {"name":"MiniMax-M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.7","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true},
  "minimax/minimax-m3": {"name":"MiniMax-M3","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","canonical_model_id":"minimax/MiniMax-M3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":512000},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-06-01","last_updated":"2026-06-01","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "minimax/minimax-m3.1-flash": {"name":"MiniMax-M3.1-Flash","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":524288},"cost":{"input":0.075,"output":0.3,"cache_read":0.0075},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "mistralai/codestral-2508": {"name":"Codestral 25.08","description":"Mistral coding model for code completion, generation, and developer workflows","family":"codestral","canonical_model_id":"mistral/codestral-2508","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":256000,"output":204800},"cost":{"input":0.3,"output":0.9,"cache_read":0.03},"release_date":"2025-07-30","last_updated":"2025-07-30","open_weights":false},
  "mistralai/devstral-2512": {"name":"Devstral 2","description":"Mistral coding agent model for repository tasks and software engineering workflows","family":"devstral","canonical_model_id":"mistral/devstral-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":262144,"output":209715},"cost":{"input":0.4,"output":2,"cache_read":0.04},"knowledge":"2025-12","release_date":"2025-12-09","last_updated":"2025-12-09","open_weights":true},
  "mistralai/ministral-14b-2512": {"name":"Ministral 3 14B 2512","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","canonical_model_id":"mistral/ministral-14b-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":209715},"cost":{"input":0.2,"output":0.2,"cache_read":0.02},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistralai/ministral-3b-2512": {"name":"Ministral 3 3B 2512","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","canonical_model_id":"mistral/ministral-3b-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":104857},"cost":{"input":0.1,"output":0.1,"cache_read":0.01},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistralai/ministral-8b-2512": {"name":"Ministral 3 8B 2512","description":"Compact Mistral model for edge, latency-sensitive, and cost-efficient workloads","family":"ministral","canonical_model_id":"mistral/ministral-8b-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":209715},"cost":{"input":0.15,"output":0.15,"cache_read":0.015},"release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistralai/mistral-large": {"name":"Mistral Large","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","canonical_model_id":"mistral/mistral-large","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":128000,"output":102400},"cost":{"input":2,"output":6,"cache_read":0.2},"knowledge":"2024-11-30","release_date":"2024-02-26","last_updated":"2024-02-26","open_weights":false},
  "mistralai/mistral-large-2407": {"name":"Mistral Large 2407","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-2407","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":131072,"output":104857},"cost":{"input":2,"output":6,"cache_read":0.2},"knowledge":"2024-03-31","release_date":"2024-07-24","last_updated":"2024-07-24","open_weights":true},
  "mistralai/mistral-large-2512": {"name":"Mistral Large 3","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-2512","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":209715},"cost":{"input":0.5,"output":1.5,"cache_read":0.05},"knowledge":"2024-11","release_date":"2025-12-02","last_updated":"2025-12-02","open_weights":true},
  "mistralai/mistral-large-4-0": {"name":"Mistral Large 4","description":"Flagship Mistral model for advanced reasoning, coding, and multilingual work","family":"mistral-large","canonical_model_id":"mistral/mistral-large-4","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0.7,"output":2.1,"cache_read":0.07},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "mistralai/mistral-medium-3": {"name":"Mistral Medium 3","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","canonical_model_id":"mistral/mistral-medium-3","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":131072,"output":104857},"cost":{"input":0.4,"output":2,"cache_read":0.04},"knowledge":"2025-03-31","release_date":"2025-05-07","last_updated":"2025-05-07","open_weights":false},
  "mistralai/mistral-medium-3-5": {"name":"Mistral Medium 3.5","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","canonical_model_id":"mistral/mistral-medium-3-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":209715},"cost":{"input":1.5,"output":7.5},"release_date":"2026-04-30","last_updated":"2026-04-30","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "mistralai/mistral-medium-3.1": {"name":"Mistral Medium 3.1","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","canonical_model_id":"mistral/mistral-medium-3.1","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":131072,"output":104857},"cost":{"input":0.4,"output":2,"cache_read":0.04},"knowledge":"2025-06-30","release_date":"2025-08-12","last_updated":"2025-08-12","open_weights":false},
  "mistralai/mistral-nemo": {"name":"Mistral Nemo","description":"Efficient Mistral-NVIDIA open model for multilingual chat and local deployment","family":"mistral-nemo","canonical_model_id":"mistral/mistral-nemo","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.029,"output":0.03},"knowledge":"2024-07","release_date":"2024-07-18","last_updated":"2024-07-18","open_weights":true},
  "mistralai/mistral-saba": {"name":"Saba","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral","canonical_model_id":"mistral/mistral-saba","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":32768,"output":26214},"cost":{"input":0.2,"output":0.6,"cache_read":0.02},"knowledge":"2024-09-30","release_date":"2025-02-17","last_updated":"2025-02-17","open_weights":false},
  "mistralai/mistral-small-24b-instruct-2501": {"name":"Mistral Small 3","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","canonical_model_id":"mistral/mistral-small-24b-instruct-2501","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":16384},"cost":{"input":0.05,"output":0.08},"knowledge":"2023-10-31","release_date":"2025-01-30","last_updated":"2025-01-30","open_weights":true},
  "mistralai/mistral-small-2603": {"name":"Mistral Small 4","description":"Fast Mistral production model for chat, extraction, and cost-sensitive agents","family":"mistral-small","canonical_model_id":"mistral/mistral-small-2603","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":209715},"cost":{"input":0.15,"output":0.6,"cache_read":0.015},"knowledge":"2025-06","release_date":"2026-03-16","last_updated":"2026-03-16","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "mistralai/mistral-small-3.1-24b-instruct": {"name":"Mistral Small 3.1 24B","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","canonical_model_id":"mistral/mistral-small-3.1-24b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":102400},"cost":{"input":0.351,"output":0.555},"knowledge":"2023-10-31","release_date":"2025-03-17","last_updated":"2025-03-17","open_weights":true},
  "mistralai/mistral-small-3.2-24b-instruct": {"name":"Mistral Small 3.2 24B","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","canonical_model_id":"mistral/mistral-small-3.2-24b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text"],"output":["text"]},"limit":{"context":256000,"output":16384},"cost":{"input":0.075,"output":0.2},"knowledge":"2023-10-31","release_date":"2025-06-20","last_updated":"2025-06-20","open_weights":true},
  "mistralai/mixtral-8x22b-instruct": {"name":"Mixtral 8x22B Instruct","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral","canonical_model_id":"mistral/mixtral-8x22b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","pdf"],"output":["text"]},"limit":{"context":65536,"output":52428},"cost":{"input":2,"output":6,"cache_read":0.2},"knowledge":"2024-01-31","release_date":"2024-04-17","last_updated":"2024-04-17","open_weights":true},
  "mistralai/voxtral-mini-tts-2603": {"name":"Voxtral Mini TTS","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "mistralai/voxtral-small-24b-2507": {"name":"Voxtral Small 24B 2507","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"voxtral","canonical_model_id":"mistral/voxtral-small-24b-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","audio","pdf"],"output":["text"]},"limit":{"context":32768,"output":26214},"cost":{"input":0.1,"output":0.3,"cache_read":0.01},"release_date":"2025-07-15","last_updated":"2025-07-15","open_weights":true},
  "moondream/moondream3-preview": {"name":"Moondream 3 Preview","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":32000,"output":4096},"cost":{"input":0.3,"output":2.5},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "moondream/moondream3.1-9b-a2b": {"name":"Moondream 3.1","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":32000,"output":4096},"cost":{"input":0.3,"output":2.5},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "moonshotai/kimi-k2": {"name":"Kimi K2 0711","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":98304},"cost":{"input":0.57,"output":2.3},"knowledge":"2024-12-31","release_date":"2025-07-11","last_updated":"2025-07-11","open_weights":true},
  "moonshotai/kimi-k2-0905": {"name":"Kimi K2 Instruct 0905","description":"Kimi model for long-context chat, coding, and agentic reasoning","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2-0905","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":98304},"cost":{"input":0.6,"output":2.5},"release_date":"2025-09-05","last_updated":"2025-09-05","open_weights":true},
  "moonshotai/kimi-k2-thinking": {"name":"Kimi K2 Thinking","description":"Thinking Kimi model for slower research passes, planning, and hard technical questions","family":"kimi-thinking","canonical_model_id":"moonshotai/kimi-k2-thinking","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":98304},"cost":{"input":0.6,"output":2.5,"cache_read":0.15},"knowledge":"2024-08","release_date":"2025-11-06","last_updated":"2025-11-06","open_weights":true},
  "moonshotai/kimi-k2.5": {"name":"Kimi K2.5","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.5,"output":2.5,"cache_read":0.15},"knowledge":"2025-01","release_date":"2026-01-27","last_updated":"2026-01-27","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "moonshotai/kimi-k2.6": {"name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.65,"output":3.41,"cache_read":0.15},"knowledge":"2025-01","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true,"rig":{"pinned":["reasoning_options"],"reasoning_field":"reasoning_content"}},
  "moonshotai/kimi-k2.7-code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.6,"output":3.5,"cache_read":0.15},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true},
  "moonshotai/kimi-k3": {"name":"Kimi K3","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.192,"output":15,"cache_read":0.182},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "moonshotai/kimi-k3.1-pro": {"name":"Kimi K3.1 Pro","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3.1-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":2,"output":10,"cache_read":0.2},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "morph/morph-v3-fast": {"name":"Morph V3 Fast","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"morph","canonical_model_id":"morph/morph-v3-fast","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.8,"output":1.2},"release_date":"2025-07-07","last_updated":"2025-07-07","open_weights":false},
  "morph/morph-v3-large": {"name":"Morph V3 Large","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"morph","canonical_model_id":"morph/morph-v3-large","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":0.9,"output":1.9},"release_date":"2025-07-07","last_updated":"2025-07-07","open_weights":false},
  "nex-agi/nex-n2.5-mini": {"name":"Nex-N2.5-Mini","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"agi","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.025,"output":0.1,"cache_read":0.0025},"release_date":"2026-09-08","last_updated":"2026-09-08","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nex-agi/nex-n2.5-pro": {"name":"Nex-N2.5-Pro","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"agi","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.075,"output":0.25,"cache_read":0.015},"release_date":"2026-09-08","last_updated":"2026-09-08","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nousresearch/hermes-3-llama-3.1-405b": {"name":"Hermes 3 405B Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"nousresearch","canonical_model_id":"nousresearch/hermes-3-llama-3.1-405b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":1,"output":1},"knowledge":"2023-12-31","release_date":"2024-08-16","last_updated":"2024-08-16","open_weights":true,"status":"deprecated"},
  "nousresearch/hermes-3-llama-3.1-70b": {"name":"Hermes 3 70B Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"nousresearch","canonical_model_id":"nousresearch/hermes-3-llama-3.1-70b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.7,"output":0.7},"knowledge":"2023-12-31","release_date":"2024-08-18","last_updated":"2024-08-18","open_weights":true,"status":"deprecated"},
  "nousresearch/hermes-4-405b": {"name":"Hermes 4 405B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"hermes","canonical_model_id":"nousresearch/hermes-4-405b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":1,"output":3,"cache_read":0.2},"knowledge":"2024-08-31","release_date":"2025-08-26","last_updated":"2025-08-26","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3-nano-30b-a3b": {"name":"Nemotron 3 Nano 30B A3B","description":"Small Nemotron 3 MoE for efficient coding, math, and long-context agents","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-nano-30b-a3b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.06,"output":0.24,"cache_read":0.012},"knowledge":"2025-06-25","release_date":"2025-12-15","last_updated":"2025-12-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3-nano-omni-30b-a3b-reasoning:free": {"name":"Nemotron 3 Nano Omni (free)","description":"Open Nemotron omni model combining reasoning with text, vision, and audio","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-nano-omni-30b-a3b-reasoning","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0,"output":0},"release_date":"2026-04-28","last_updated":"2026-04-28","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3-super-120b-a12b": {"name":"Nemotron 3 Super 120B A12B","description":"Nemotron middle tier for collaborative agents and high-volume reasoning workloads","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-super-120b-a12b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0.3,"output":2.4,"cache_read":0.15},"knowledge":"2025-06","release_date":"2026-03-11","last_updated":"2026-03-11","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3-ultra-550b-a55b": {"name":"Nemotron 3 Ultra 550B A55B","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-ultra-550b-a55b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":182520},"cost":{"input":0.6,"output":2.4,"cache_read":0.12},"knowledge":"2025-09","release_date":"2026-06-04","last_updated":"2026-06-04","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3-ultra-550b-a55b:free": {"name":"Nemotron 3 Ultra (free)","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-ultra-550b-a55b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0,"output":0},"knowledge":"2025-09","release_date":"2026-06-04","last_updated":"2026-06-04","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3.5-content-safety": {"name":"Nemotron 3.5 Content Safety","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-content-safety","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":0.2,"output":0.2},"release_date":"2026-06-02","last_updated":"2026-06-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3.5-content-safety:free": {"name":"Nemotron 3.5 Content Safety (free)","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-content-safety","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0,"output":0},"release_date":"2026-06-02","last_updated":"2026-06-02","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3.5-lightning": {"name":"Nemotron 3.5 Lightning 30B A3B","description":"Nemotron model for efficient reasoning, coding, and specialized AI agents","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-lightning","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.07,"output":0.2,"cache_read":0.04},"knowledge":"2025-09","release_date":"2026-08-11","last_updated":"2026-08-11","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3.5-lightning:free": {"name":"Nemotron 3.5 Lightning (free)","description":"Nemotron model for efficient reasoning, coding, and specialized AI agents","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-lightning","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0,"output":0},"knowledge":"2025-09","release_date":"2026-08-11","last_updated":"2026-08-11","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3.5-super-vl": {"name":"Nemotron 3.5 Super VL","description":"Nemotron multimodal model for visual reasoning and agentic AI workflows","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-super-vl-120b-a12b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":235929},"cost":{"input":0.15,"output":0.5,"cache_read":0.02},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "nvidia/nemotron-3.5-super-vl:free": {"name":"Nemotron 3.5 Super VL (free)","description":"Nemotron multimodal model for visual reasoning and agentic AI workflows","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-super-vl-120b-a12b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0,"output":0},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-3.5-turbo-0613": {"name":"GPT-3.5 Turbo (older v0613)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":4095,"output":3685},"cost":{"input":1,"output":2},"knowledge":"2021-09-30","release_date":"2024-01-25","last_updated":"2024-01-25","open_weights":false},
  "openai/gpt-3.5-turbo-16k": {"name":"GPT-3.5 Turbo 16k","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":16385,"output":4096},"cost":{"input":3,"output":4},"knowledge":"2021-09-30","release_date":"2023-08-28","last_updated":"2023-08-28","open_weights":false},
  "openai/gpt-4": {"name":"GPT-4","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-4","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8191,"output":4096},"cost":{"input":30,"output":60},"knowledge":"2023-11","release_date":"2023-11-06","last_updated":"2024-04-09","open_weights":false},
  "openai/gpt-4.1": {"name":"GPT-4.1","description":"Long-lived GPT workhorse for coding, instruction following, and production apps","family":"gpt","canonical_model_id":"openai/gpt-4.1","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":2,"output":8,"cache_read":0.5},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false},
  "openai/gpt-4.1-mini": {"name":"GPT-4.1 mini","description":"Affordable GPT-4.1 lane for fast coding help and structured extraction","family":"gpt-mini","canonical_model_id":"openai/gpt-4.1-mini","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1047576,"output":32768},"cost":{"input":0.4,"output":1.6,"cache_read":0.1},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false},
  "openai/gpt-4.1-nano": {"name":"GPT-4.1 nano","description":"Tiny GPT-4.1 option for classification, routing, and very high-volume tasks","family":"gpt-nano","canonical_model_id":"openai/gpt-4.1-nano","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text","pdf"],"output":["text"]},"limit":{"context":1047576,"output":942818},"cost":{"input":0.1,"output":0.4,"cache_read":0.03},"knowledge":"2024-04","release_date":"2025-04-14","last_updated":"2025-04-14","open_weights":false},
  "openai/gpt-4o": {"name":"GPT-4o","description":"Omni-era GPT for multimodal chat, practical coding, and general assistants","family":"gpt","canonical_model_id":"openai/gpt-4o","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-08-06","open_weights":false},
  "openai/gpt-4o-2024-05-13": {"name":"GPT-4o (2024-05-13)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-4o-2024-05-13","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":5,"output":15},"knowledge":"2023-09","release_date":"2024-05-13","last_updated":"2024-05-13","open_weights":false},
  "openai/gpt-4o-2024-08-06": {"name":"GPT-4o (2024-08-06)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-4o-2024-08-06","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-08-06","last_updated":"2024-08-06","open_weights":false},
  "openai/gpt-4o-2024-11-20": {"name":"GPT-4o (2024-11-20)","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-4o-2024-11-20","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-11-20","last_updated":"2024-11-20","open_weights":false},
  "openai/gpt-4o-mini": {"name":"GPT-4o mini","description":"Small omni GPT for cheap multimodal assistance and production-scale traffic","family":"gpt-mini","canonical_model_id":"openai/gpt-4o-mini","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075},"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","open_weights":false},
  "openai/gpt-4o-mini-2024-07-18": {"name":"GPT-4o-mini (2024-07-18)","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","canonical_model_id":"openai/gpt-4o-mini","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075},"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","open_weights":false},
  "openai/gpt-4o-mini-transcribe": {"name":"GPT-4o mini Transcribe","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "openai/gpt-4o-mini-tts-2025-12-15": {"name":"GPT-4o mini TTS","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "openai/gpt-4o-transcribe": {"name":"GPT-4o Transcribe","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "openai/gpt-5": {"name":"GPT-5","description":"Original GPT-5 workhorse for reasoning, coding, writing, and tool workflows","family":"gpt","canonical_model_id":"openai/gpt-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5-image": {"name":"GPT-5 Image","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt","type":"image","canonical_model_id":"openai/gpt-5-image","reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text","pdf"],"output":["image","text"]},"limit":{"context":400000,"output":128000},"cost":{"input":10,"output":10,"cache_read":1.25},"knowledge":"2024-10-01","release_date":"2025-10-14","last_updated":"2025-10-14","open_weights":false},
  "openai/gpt-5-image-mini": {"name":"GPT-5 Image Mini","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt","type":"image","canonical_model_id":"openai/gpt-5-image-mini","reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["image","text"]},"limit":{"context":400000,"output":128000},"cost":{"input":2.5,"output":2,"cache_read":0.25},"release_date":"2025-10-16","last_updated":"2025-10-16","open_weights":false},
  "openai/gpt-5-mini": {"name":"GPT-5 Mini","description":"Small GPT-5 for responsive agents, coding help, and everyday automation","family":"gpt-mini","canonical_model_id":"openai/gpt-5-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.025},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5-nano": {"name":"GPT-5 Nano","description":"Tiny GPT-5 lane for routing, extraction, classification, and bulk jobs","family":"gpt-nano","canonical_model_id":"openai/gpt-5-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.05,"output":0.4,"cache_read":0.005},"knowledge":"2024-05-30","release_date":"2025-08-07","last_updated":"2025-08-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5-pro": {"name":"GPT-5 Pro","description":"Higher-accuracy GPT-5 tier for tough analysis, coding reviews, and planning","family":"gpt-pro","canonical_model_id":"openai/gpt-5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["image","text","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":15,"output":120},"knowledge":"2024-09-30","release_date":"2025-10-06","last_updated":"2025-10-06","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.1": {"name":"GPT-5.1","description":"Sharper GPT-5 generation for coding, product work, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-5.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["image","text","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.1-codex": {"name":"GPT-5.1 Codex","description":"Codex GPT for repository edits, code review, and practical software agents","family":"gpt-codex","canonical_model_id":"openai/gpt-5.1-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.13},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.1-codex-max": {"name":"GPT-5.1 Codex Max","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.1-codex-max","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.25,"output":10,"cache_read":0.125},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.1-codex-mini": {"name":"GPT-5.1 Codex mini","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.1-codex-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.25,"output":2,"cache_read":0.03},"knowledge":"2024-09-30","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.2": {"name":"GPT-5.2","description":"Reliable GPT generation for broad coding, writing, and tool-assisted product work","family":"gpt","canonical_model_id":"openai/gpt-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.2-codex": {"name":"GPT-5.2 Codex","description":"Code-specialist GPT for repository edits, reviews, and long-running software agents","family":"gpt-codex","canonical_model_id":"openai/gpt-5.2-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.2-pro": {"name":"GPT-5.2 Pro","description":"Higher-accuracy GPT-5.2 variant for tougher reasoning and review workflows","family":"gpt-pro","canonical_model_id":"openai/gpt-5.2-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["image","text","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":21,"output":168},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.3-codex": {"name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.3-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.4": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","canonical_model_id":"openai/gpt-5.4","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.4-image-2": {"name":"GPT-5.4 Image 2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gpt","type":"image","canonical_model_id":"openai/gpt-5.4-image-2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":false,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["image","text","pdf"],"output":["image","text"]},"limit":{"context":272000,"output":128000},"cost":{"input":8,"output":15,"cache_read":2},"release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.4-mini": {"name":"GPT-5.4 mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","canonical_model_id":"openai/gpt-5.4-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.4-nano": {"name":"GPT-5.4 nano","description":"Cheapest GPT-5.4 lane for simple routing, extraction, and bulk automation","family":"gpt-nano","canonical_model_id":"openai/gpt-5.4-nano","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.2,"output":1.25,"cache_read":0.02},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.4-pro": {"name":"GPT-5.4 Pro","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","canonical_model_id":"openai/gpt-5.4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.5": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","canonical_model_id":"openai/gpt-5.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.5-pro": {"name":"GPT-5.5 Pro","description":"Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding","family":"gpt-pro","canonical_model_id":"openai/gpt-5.5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.6-luna": {"name":"GPT-5.6 Luna","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.6-luna-pro": {"name":"GPT-5.6 Luna Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.6-sol": {"name":"GPT-5.6 Sol","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.6-sol-pro": {"name":"GPT-5.6 Sol Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.6-terra": {"name":"GPT-5.6 Terra","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-5.6-terra-pro": {"name":"GPT-5.6 Terra Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6-astra-pro": {"name":"GPT-6 Astra Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6-luna": {"name":"GPT-6 Luna","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6-luna-pro": {"name":"GPT-6 Luna Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6-sol": {"name":"GPT-6 Sol","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6-sol-pro": {"name":"GPT-6 Sol Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6.1-sol": {"name":"GPT-6.1 Sol","description":"Near-Astra performance for complex coding, computer use, and professional work at a lower cost.","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6.1-sol-pro": {"name":"GPT-6.1 Sol Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6.1-terra": {"name":"GPT-6.1 Terra","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-6.1-terra-pro": {"name":"GPT-6.1 Terra Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-audio": {"name":"GPT-Audio","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gpt","type":"chat","canonical_model_id":"openai/gpt-audio","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","audio"],"output":["text","audio"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"input_audio":32,"output_audio":64},"knowledge":"2023-10-01","release_date":"2025-08-28","last_updated":"2025-08-28","open_weights":false},
  "openai/gpt-audio-mini": {"name":"GPT-Audio Mini","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gpt-mini","type":"chat","canonical_model_id":"openai/gpt-audio-mini","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","audio"],"output":["text","audio"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.6,"output":2.4,"input_audio":0.6,"output_audio":2.4},"knowledge":"2023-10-01","release_date":"2025-10-06","last_updated":"2025-10-06","open_weights":false},
  "openai/gpt-chat-latest": {"name":"GPT Chat Latest","description":"Chat-tuned GPT model for conversational assistance, writing, and tool workflows","family":"gpt","reasoning":false,"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":400000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5},"release_date":"2026-05-05","last_updated":"2026-05-05","open_weights":false},
  "openai/gpt-oss-120b": {"name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.15,"output":0.6,"cache_read":0.075},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-oss-20b": {"name":"GPT OSS 20B","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.018,"output":0.09,"cache_read":0.009},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "openai/gpt-oss-safeguard-20b": {"name":"GPT OSS Safeguard 20B","description":"Safety model for policy screening, moderation, and risk-aware routing workflows","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-safeguard-20b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":65536},"cost":{"input":0.075,"output":0.3,"cache_read":0.0375},"release_date":"2025-10-29","last_updated":"2025-10-29","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "openai/o3": {"name":"o3","description":"Deliberate o-series reasoner for hard math, coding, and multi-step analysis","family":"o","canonical_model_id":"openai/o3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":2,"output":8,"cache_read":0.5},"knowledge":"2024-05","release_date":"2025-04-16","last_updated":"2025-04-16","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/o3-pro": {"name":"o3-pro","description":"High-effort o3 tier for difficult technical reasoning and careful answers","family":"o-pro","canonical_model_id":"openai/o3-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","pdf","image"],"output":["text"]},"limit":{"context":200000,"output":100000},"cost":{"input":20,"output":80},"knowledge":"2024-05","release_date":"2025-06-10","last_updated":"2025-06-10","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "openai/whisper-1": {"name":"Whisper","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "openai/whisper-large-v3": {"name":"Whisper large v3","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "openai/whisper-large-v3-turbo": {"name":"Whisper large v3 turbo","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "openrouter/auto": {"name":"Auto Router","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"auto","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","pdf","video"],"output":["text","image"]},"limit":{"context":2000000,"output":2000000},"release_date":"2023-11-08","last_updated":"2023-11-08","open_weights":false},
  "openrouter/bodybuilder": {"name":"Body Builder (beta)","description":"Preview model for early access evaluation, prototyping, and compatibility testing","reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":128000},"release_date":"2025-12-05","last_updated":"2025-12-05","open_weights":false},
  "openrouter/free": {"name":"Free Models Router","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"input":200000,"output":8000},"cost":{"input":0,"output":0},"release_date":"2026-02-01","last_updated":"2026-02-01","open_weights":false},
  "openrouter/fusion": {"name":"Fusion","description":"General-purpose chat model for instruction following, writing, and analysis","reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":128000},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":false},
  "openrouter/pareto-code": {"name":"Pareto Code Router","description":"Coding model for repository understanding, refactors, and agentic engineering tasks","reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":2000000,"output":200000},"release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false},
  "particle-ai/particle-1": {"name":"Particle-1","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.05,"output":0.5,"cache_read":0.003},"release_date":"2026-10-28","last_updated":"2026-10-28","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "perceptron/perceptron-mk1": {"name":"Perceptron Mk1","description":"Multimodal reasoning model for visual analysis, planning, and tool use","canonical_model_id":"perceptron/perceptron-mk1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":32768,"output":8192},"cost":{"input":0.15,"output":1.5},"release_date":"2026-05-12","last_updated":"2026-05-12","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "perceptron/perceptron-mk1.5": {"name":"Perceptron Mk1.5","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio"],"output":["text"]},"limit":{"context":36864,"output":8192},"cost":{"input":0.15,"output":1.5},"release_date":"2026-09-25","last_updated":"2026-09-25","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "perplexity/sonar": {"name":"Sonar","description":"Sonar search model for current answers, retrieval, and citation-backed chat","family":"sonar","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":127072,"output":114364},"cost":{"input":1,"output":1},"release_date":"2025-01-27","last_updated":"2025-01-27","open_weights":false},
  "perplexity/sonar-deep-research": {"name":"Sonar Deep Research","description":"Sonar search model for current answers, retrieval, and citation-backed chat","family":"sonar-deep-research","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":115200},"cost":{"input":2,"output":8,"reasoning":3},"release_date":"2025-03-07","last_updated":"2025-03-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "perplexity/sonar-pro": {"name":"Sonar Pro","description":"Advanced Sonar search model for deeper research and cited synthesis","family":"sonar-pro","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":8000},"cost":{"input":3,"output":15},"release_date":"2025-03-07","last_updated":"2025-03-07","open_weights":false},
  "perplexity/sonar-pro-search": {"name":"Sonar Pro Search","description":"Advanced Sonar search model for deeper research and cited synthesis","family":"sonar-pro","reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":8000},"cost":{"input":3,"output":15},"release_date":"2025-10-30","last_updated":"2025-10-30","open_weights":false},
  "perplexity/sonar-reasoning-pro": {"name":"Sonar Reasoning Pro","description":"Web-grounded reasoning model for multi-step research and cited answers","family":"sonar-reasoning","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":115200},"cost":{"input":2,"output":8},"release_date":"2025-03-07","last_updated":"2025-03-07","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "poolside/laguna-s-2.1": {"name":"Laguna S 2.1","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"laguna","canonical_model_id":"poolside/laguna-s-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.09,"output":0.18,"cache_read":0.009},"release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "poolside/laguna-s-2.1:free": {"name":"Laguna S 2.1 (free)","description":"Free provider route for experiments, demos, and cost-sensitive chat workloads","family":"laguna","canonical_model_id":"poolside/laguna-s-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0},"release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "poolside/laguna-xs-2.1": {"name":"Laguna XS 2.1","description":"Agentic coding model from Poolside in the XS size class for local deployment","family":"laguna","canonical_model_id":"poolside/laguna-xs-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.06,"output":0.12,"cache_read":0.03},"release_date":"2026-07-02","last_updated":"2026-07-02","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "poolside/laguna-xs-2.1:free": {"name":"Laguna XS 2.1 (free)","description":"Free provider route for experiments, demos, and cost-sensitive chat workloads","family":"laguna","canonical_model_id":"poolside/laguna-xs-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0,"output":0},"release_date":"2026-07-02","last_updated":"2026-07-02","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "prism-ml/ternary-bonsai-2-27b": {"name":"Ternary Bonsai 2 27B","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.075,"output":0.5,"cache_read":0.0375},"release_date":"2026-09-18","last_updated":"2026-09-18","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "prism-ml/ternary-bonsai-2-9b": {"name":"Ternary Bonsai 2 9B","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.08,"output":0.13,"cache_read":0.04},"release_date":"2026-10-16","last_updated":"2026-10-16","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen-2.5-72b-instruct": {"name":"Qwen2.5 72B Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":16384},"cost":{"input":0.36,"output":0.4},"knowledge":"2024-06-30","release_date":"2024-09-19","last_updated":"2024-09-19","open_weights":true},
  "qwen/qwen-2.5-7b-instruct": {"name":"Qwen2.5 7B Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":29491},"cost":{"input":0.1,"output":0.2},"knowledge":"2024-06-30","release_date":"2024-10-16","last_updated":"2024-10-16","open_weights":true},
  "qwen/qwen-2.5-coder-32b-instruct": {"name":"Qwen2.5-Coder-32B-Instruct","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","canonical_model_id":"alibaba/qwen2.5-coder-32b-instruct","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":29491},"cost":{"input":0.66,"output":1},"release_date":"2024-11-12","last_updated":"2024-11-12","open_weights":true},
  "qwen/qwen-plus": {"name":"Qwen Plus","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen-plus","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":0.26,"output":0.78,"cache_read":0.052,"cache_write":0.325,"tiers":[{"input":0.78,"output":2.34,"cache_read":0.156,"cache_write":0.975,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.78,"output":2.34,"cache_read":0.156,"cache_write":0.975}},"knowledge":"2024-04","release_date":"2024-01-25","last_updated":"2025-09-11","open_weights":false},
  "qwen/qwen2.5-vl-72b-instruct": {"name":"Qwen2.5-VL 72B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen2-5-vl-72b-instruct","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":115200},"cost":{"input":0.8,"output":1,"cache_read":0.4},"knowledge":"2024-04","release_date":"2025-01-26","last_updated":"2025-01-26","open_weights":true},
  "qwen/qwen3-14b": {"name":"Qwen3 14B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":40960,"output":16384},"cost":{"input":0.12,"output":0.24},"knowledge":"2025-03-31","release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3-235b-a22b-2507": {"name":"Qwen3 235B A22B Instruct 2507","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-instruct-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":0.09,"output":0.55},"release_date":"2025-07-21","last_updated":"2025-07-21","open_weights":true},
  "qwen/qwen3-235b-a22b-thinking-2507": {"name":"Qwen3 235B-A22B Thinking 2507","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-thinking-2507","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.45,"output":3.5},"release_date":"2025-07-25","last_updated":"2025-07-25","open_weights":true},
  "qwen/qwen3-30b-a3b": {"name":"Qwen3 30B A3B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-30b-a3b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":40960,"output":16384},"cost":{"input":0.12,"output":0.5},"release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3-30b-a3b-instruct-2507": {"name":"Qwen3 30B-A3B Instruct 2507","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-30b-a3b-instruct-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.1,"output":0.3,"cache_read":0.1},"release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":true},
  "qwen/qwen3-32b": {"name":"Qwen3 32B","description":"Dense open Qwen model for self-hosted chat, reasoning, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3-32b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.08,"output":0.28},"knowledge":"2025-04","release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3-coder": {"name":"Qwen3 Coder 480B A35B","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-480b-a35b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.3,"output":1,"cache_read":0.1},"knowledge":"2025-04","release_date":"2025-07-22","last_updated":"2025-07-22","open_weights":true},
  "qwen/qwen3-coder-30b-a3b-instruct": {"name":"Qwen3-Coder 30B-A3B Instruct","description":"Smaller Qwen coder for efficient local agents and repo-level fixes","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-30b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.07,"output":0.28},"knowledge":"2025-04","release_date":"2025-07-31","last_updated":"2025-07-31","open_weights":true},
  "qwen/qwen3-coder-flash": {"name":"Qwen3 Coder Flash","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-flash","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.195,"output":0.975,"cache_read":0.039,"cache_write":0.24375,"tiers":[{"input":0.325,"output":1.625,"cache_read":0.065,"cache_write":0.40625,"tier":{"type":"context","size":32000}},{"input":0.52,"output":2.6,"cache_read":0.104,"cache_write":0.65,"tier":{"type":"context","size":128000}}]},"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":false},
  "qwen/qwen3-coder-next": {"name":"Qwen3 Coder Next","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-next","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.12,"output":0.8,"cache_read":0.07},"knowledge":"2025-09","release_date":"2026-02-03","last_updated":"2026-02-03","open_weights":true},
  "qwen/qwen3-next-80b-a3b-instruct": {"name":"Qwen3-Next 80B-A3B Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-next-80b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":0.09,"output":1.1},"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","open_weights":true},
  "qwen/qwen3-vl-235b-a22b-instruct": {"name":"Qwen3 VL 235B A22B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3-vl-235b-a22b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.21,"output":1.9,"cache_read":0.1},"knowledge":"2025-03-31","release_date":"2025-09-23","last_updated":"2025-09-23","open_weights":true},
  "qwen/qwen3-vl-30b-a3b-instruct": {"name":"Qwen3 VL 30B A3B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3-vl-30b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":0.15,"output":0.6},"release_date":"2025-10-04","last_updated":"2025-10-04","open_weights":true},
  "qwen/qwen3-vl-30b-a3b-thinking": {"name":"Qwen3 VL 30B A3B Thinking","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.29,"output":1},"knowledge":"2025-03-31","release_date":"2025-10-06","last_updated":"2025-10-06","open_weights":true},
  "qwen/qwen3-vl-8b-instruct": {"name":"Qwen3 VL 8B Instruct","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.25,"output":0.75,"cache_read":0.12},"release_date":"2025-10-14","last_updated":"2025-10-14","open_weights":true},
  "qwen/qwen3.5-122b-a10b": {"name":"Qwen3.5 122B-A10B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-122b-a10b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.26,"output":2.08},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.5-27b": {"name":"Qwen3.5 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":81920},"cost":{"input":0.26,"output":2.6},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.5-35b-a3b": {"name":"Qwen3.5 35B-A3B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-35b-a3b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":81920},"cost":{"input":0.14,"output":1,"cache_read":0.05},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.5-397b-a17b": {"name":"Qwen3.5 397B-A17B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-397b-a17b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.5,"output":3.6,"cache_read":0.3},"release_date":"2026-02-15","last_updated":"2026-02-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.5-9b": {"name":"Qwen3.5 9B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.5-9b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.17,"output":0.25},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.5-flash-02-23": {"name":"Qwen3.5-Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.065,"output":0.26},"release_date":"2026-02-25","last_updated":"2026-02-25","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.5-plus-02-15": {"name":"Qwen3.5 Plus 2026-02-15","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.26,"output":1.56,"tiers":[{"input":0.325,"output":1.95,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.325,"output":1.95}},"knowledge":"2025-04","release_date":"2026-02-16","last_updated":"2026-02-16","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.5-plus-20260420": {"name":"Qwen3.5 Plus 2026-04-20","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.3,"output":1.8,"cache_write":0.375,"tiers":[{"input":0.375,"output":2.25,"cache_write":0.46875,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.375,"output":2.25,"cache_write":0.46875}},"release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.6-27b": {"name":"Qwen3.6 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.6-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":81920},"cost":{"input":0.32,"output":3.2},"release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.6-35b-a3b": {"name":"Qwen3.6 35B-A3B","description":"Open multimodal Qwen MoE for local agents that need vision, audio, and code","family":"qwen","canonical_model_id":"alibaba/qwen3.6-35b-a3b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.1,"output":0.9,"cache_read":0.025},"release_date":"2026-04-17","last_updated":"2026-04-17","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.6-flash": {"name":"Qwen3.6 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.6","canonical_model_id":"alibaba/qwen3.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.1875,"output":1.125,"cache_write":0.234375,"tiers":[{"input":0.75,"output":3,"cache_write":0.9375,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.75,"output":3,"cache_write":0.9375}},"release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.6-plus": {"name":"Qwen3.6 Plus","description":"Earlier Qwen multimodal workhorse for million-token agent and document tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.6-plus","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.325,"output":1.95,"cache_write":0.40625,"tiers":[{"input":1.3,"output":3.9,"cache_write":1.625,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":1.3,"output":3.9,"cache_write":1.625}},"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.7-flash": {"name":"Qwen3.7 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"input":991808,"output":65536},"cost":{"input":0.03,"output":0.13,"cache_read":0.006,"cache_write":0.038,"tiers":[{"input":0.1,"output":0.4,"cache_read":0.02,"cache_write":0.125,"tier":{"type":"context","size":32000}},{"input":0.2,"output":0.8,"cache_read":0.04,"cache_write":0.25,"tier":{"type":"context","size":256000}}]},"release_date":"2026-07-15","last_updated":"2026-07-15","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.7-max": {"name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.7-max","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.475,"output":4.425,"cache_read":0.295,"cache_write":1.84375},"release_date":"2026-05-21","last_updated":"2026-05-21","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.7-plus": {"name":"Qwen3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3.7-plus","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.32,"output":1.28,"cache_read":0.064,"cache_write":0.4,"tiers":[{"input":0.96,"output":3.84,"cache_read":0.192,"cache_write":1.2,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":0.96,"output":3.84,"cache_read":0.192,"cache_write":1.2}},"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.8-2.4t-a95b": {"name":"Qwen3.8 2.4T A95B","description":"Open-weight sparse MoE (2.4T total, 95B active), the open-weight twin of Qwen3.8 Max for coding, research, complex reasoning, and agentic workflows","family":"qwen","canonical_model_id":"alibaba/qwen3.8-2.4t-a95b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25},"release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.8-27b": {"name":"Qwen3.8 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":235929},"cost":{"input":0.15,"output":1.875,"cache_read":0.0375},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.8-flash": {"name":"Qwen3.8 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.47,"cache_read":0.016},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.8-max-0902": {"name":"Qwen3.8 Max 0902","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-max-0902","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":2,"output":6,"cache_read":0.25},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.8-max-prime": {"name":"Qwen 3.8 Max Prime","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen3.8-max","canonical_model_id":"alibaba/qwen3.8-max-prime","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":4,"output":12,"cache_read":0.5},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwen3.8-omni-flash": {"name":"Qwen3.8 Omni Flash","description":"Qwen omni model for text, vision, audio, and multimodal agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-omni-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.47,"cache_read":0.016},"release_date":"2026-09-17","last_updated":"2026-09-17","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "qwen/qwq-32b": {"name":"QwQ 32B","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "recursive/flint-v1.0:free": {"name":"Flint (free)","description":"Free provider route for experiments, demos, and cost-sensitive chat workloads","family":"flint","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0,"output":0},"release_date":"2026-10-21","last_updated":"2026-10-21","open_weights":true},
  "reflection/beam-1.0-501b-a23b": {"name":"Beam 1.0 501B A23B","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"beam","canonical_model_id":"reflection/beam-1.0-501b-a23b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.25,"output":1,"cache_read":0.05},"knowledge":"2026-06-30","release_date":"2026-10-05","last_updated":"2026-10-16","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "rekaai/reka-edge": {"name":"Reka Edge","description":"Multimodal model for analyzing text, images, documents, and rich media","family":"reka","canonical_model_id":"rekaai/reka-edge","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text","video"],"output":["text"]},"limit":{"context":16384,"output":14745},"cost":{"input":0.1,"output":0.1},"release_date":"2026-03-11","last_updated":"2026-03-11","open_weights":true},
  "rekaai/reka-flash-3": {"name":"Reka Flash 3","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"reka","canonical_model_id":"rekaai/reka-flash-3","reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":65536,"output":58982},"cost":{"input":0.1,"output":0.2},"knowledge":"2025-01-31","release_date":"2025-03-10","last_updated":"2025-03-10","open_weights":true},
  "relace/relace-apply-3": {"name":"Relace Apply 3","description":"General-purpose chat model for instruction following, writing, and analysis","canonical_model_id":"relace/relace-apply-3","reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":128000},"cost":{"input":0.85,"output":1.25},"release_date":"2025-09-26","last_updated":"2025-09-26","open_weights":false},
  "relace/relace-search": {"name":"Relace Search","description":"Tool-capable chat model for instruction following and agentic application workflows","canonical_model_id":"relace/relace-search","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":128000},"cost":{"input":1,"output":3},"release_date":"2025-12-08","last_updated":"2025-12-08","open_weights":false},
  "sakana/fugu-max": {"name":"Fugu Max","description":"Multi-agent model for routing expert agents across complex analytical tasks","family":"fugu","canonical_model_id":"sakana/fugu-max","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":6,"cache_read":0.25},"release_date":"2026-09-11","last_updated":"2026-09-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "sakana/fugu-ultra": {"name":"Fugu Ultra","description":"Quality-first multi-agent model for hard research, analysis, and competitions","family":"fugu","canonical_model_id":"sakana/fugu-ultra","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"release_date":"2026-06-15","last_updated":"2026-06-15","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "sakana/fugu-ultra-v2": {"name":"Fugu Ultra v2","description":"Quality-first multi-agent model for hard research, analysis, and competitions","family":"fugu","canonical_model_id":"sakana/fugu-ultra-v2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"knowledge":"2026-08-28","release_date":"2026-09-11","last_updated":"2026-09-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "sakana/sakana-namazu": {"name":"Sakana Namazu","description":"Multi-agent model for routing expert agents across complex analytical tasks","family":"sakana-namazu","canonical_model_id":"sakana/sakana-namazu","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.95,"output":4,"cache_read":0.15},"release_date":"2026-08-03","last_updated":"2026-08-03","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "sao10k/l3-lunaris-8b": {"name":"Llama 3 8B Lunaris","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"sao10k/l3-lunaris-8b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":7372},"cost":{"input":0.04,"output":0.05},"knowledge":"2023-12-31","release_date":"2024-08-13","last_updated":"2024-08-13","open_weights":true},
  "sao10k/l3.1-euryale-70b": {"name":"Llama 3.1 Euryale 70B v2.2","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"sao10k/l3.1-euryale-70b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":16384},"cost":{"input":0.85,"output":0.85},"knowledge":"2023-12-31","release_date":"2024-08-28","last_updated":"2024-08-28","open_weights":true},
  "scaledown/extract": {"name":"Extract","description":"Multimodal model for analyzing text, images, documents, and rich media","reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":900000},"cost":{"input":0.04,"output":0},"release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "scaledown/summarize": {"name":"Summarize","description":"Multimodal model for analyzing text, images, documents, and rich media","reasoning":false,"tool_call":false,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":4000},"cost":{"input":0.04,"output":0},"release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "stepfun/step-3.5-flash": {"name":"Step 3.5 Flash","description":"StepFun flash lane for quick multimodal reasoning and coding assistance","family":"step","canonical_model_id":"stepfun/step-3.5-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.1,"output":0.3},"knowledge":"2025-01","release_date":"2026-01-29","last_updated":"2026-02-13","open_weights":true},
  "stepfun/step-3.7-flash": {"name":"Step 3.7 Flash","description":"Newer StepFun flash model for faster agents, coding, and multimodal prompts","canonical_model_id":"stepfun/step-3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"input":256000,"output":230400},"cost":{"input":0.2,"output":1.15,"cache_read":0.04},"knowledge":"2026-03-01","release_date":"2026-05-29","last_updated":"2026-05-29","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "stepfun/step-5-preview": {"name":"Step 5 Preview","description":"StepFun flash model for efficient multimodal reasoning, coding, and tool use","canonical_model_id":"stepfun/step-5-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"input":1000000,"output":64000},"cost":{"input":1,"output":2.7,"cache_read":0.1},"release_date":"2026-09-16","last_updated":"2026-09-20","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "tencent/hunyuan-a13b-instruct": {"name":"Hunyuan A13B Instruct","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"hunyuan","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":0.14,"output":0.57},"knowledge":"2025-03-31","release_date":"2025-07-08","last_updated":"2025-07-08","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "tencent/hy-mt2-1.8b": {"name":"Hy-MT2-1.8B","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy-mt2-1.8b","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":4096},"cost":{"input":0.044,"output":0.177},"release_date":"2026-05-21","last_updated":"2026-05-21","open_weights":true},
  "tencent/hy-mt2-30b-a3b": {"name":"Hy-MT2-30B-A3B","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy-mt2-30b-a3b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":4096},"cost":{"input":0.074,"output":0.295},"release_date":"2026-05-21","last_updated":"2026-05-21","open_weights":true},
  "tencent/hy-mt2-7b": {"name":"Hy-MT2-7B","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy-mt2-7b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":8192,"output":4096},"cost":{"input":0.074,"output":0.295},"release_date":"2026-05-21","last_updated":"2026-05-21","open_weights":true},
  "tencent/hy3": {"name":"Hy3","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"input":192000,"output":128000},"cost":{"input":0.0825,"output":0.33,"cache_read":0.020625},"release_date":"2026-07-06","last_updated":"2026-07-06","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "tencent/hy3-preview": {"name":"Hy3 preview","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy3-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0.18,"output":0.6,"cache_read":0.06},"release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "tencent/hy4-preview": {"name":"Hy4 preview","description":"Tencent Hy reasoning model for coding, instruction following, and agent tasks","family":"Hy","canonical_model_id":"tencent/hy4-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":64000},"cost":{"input":0.7506,"output":2.2509,"cache_read":0.0378},"release_date":"2026-08-28","last_updated":"2026-08-28","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "thedrummer/cydonia-24b-v4.1": {"name":"Cydonia 24B V4.1","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","canonical_model_id":"thedrummer/cydonia-24b-v4.1","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":117964},"cost":{"input":0.3,"output":0.5,"cache_read":0.15},"knowledge":"2024-04-30","release_date":"2025-09-27","last_updated":"2025-09-27","open_weights":true},
  "thedrummer/skyfall-36b-v2": {"name":"Skyfall 36B V2","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","canonical_model_id":"thedrummer/skyfall-36b-v2","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":29491},"cost":{"input":0.55,"output":0.8,"cache_read":0.25},"knowledge":"2024-06-30","release_date":"2025-03-10","last_updated":"2025-03-10","open_weights":true},
  "thedrummer/unslopnemo-12b": {"name":"UnslopNemo 12B","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","canonical_model_id":"thedrummer/unslopnemo-12b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1024000,"output":819200},"cost":{"input":0.4,"output":0.4},"knowledge":"2024-04-30","release_date":"2024-11-08","last_updated":"2024-11-08","open_weights":true},
  "thinkingmachines/inkling": {"name":"Inkling","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"ling","canonical_model_id":"thinkingmachines/inkling","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio"],"output":["text"]},"limit":{"context":524288,"output":471859},"cost":{"input":1,"output":4.05,"cache_read":0.17},"release_date":"2026-07-15","last_updated":"2026-07-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "thinkingmachines/inkling-small": {"name":"Inkling Small","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"ling","canonical_model_id":"thinkingmachines/inkling-small","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio"],"output":["text"]},"limit":{"context":524288,"output":262144},"cost":{"input":0.45,"output":1.2,"cache_read":0.1},"release_date":"2026-07-30","last_updated":"2026-07-30","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "thinkingmachines/inkling-small:free": {"name":"Inkling Small (free)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"ling","canonical_model_id":"thinkingmachines/inkling-small","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0,"output":0},"release_date":"2026-07-30","last_updated":"2026-07-30","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "thinkingmachines/inkling:free": {"name":"Inkling (free)","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"ling","canonical_model_id":"thinkingmachines/inkling","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio"],"output":["text"]},"limit":{"context":1048576,"output":262144},"cost":{"input":0,"output":0},"release_date":"2026-07-15","last_updated":"2026-07-15","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "unbiased/pareto-26.10": {"name":"Pareto 26.10","description":"Multimodal model for analyzing text, images, documents, and rich media","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.8,"output":3.2,"cache_read":0.03},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":false},
  "unbiased/pareto-26.9": {"name":"Pareto 26.9","description":"Multimodal model for analyzing text, images, documents, and rich media","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":131072},"cost":{"input":2.5,"output":7.5,"cache_read":0.25},"release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "undi95/remm-slerp-l2-13b": {"name":"ReMM SLERP 13B","description":"Open-weight instruction model for adaptable chat and self-hosted production workloads","canonical_model_id":"undi95/remm-slerp-l2-13b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":6144,"output":5529},"cost":{"input":0.35,"output":0.65},"knowledge":"2023-06-30","release_date":"2023-07-22","last_updated":"2023-07-22","open_weights":true},
  "upstage/solar-mini4": {"name":"Solar Mini 4","description":"Efficient model for low-latency assistance, extraction, and routine automation","family":"solar-mini","canonical_model_id":"upstage/solar-mini4","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":524288,"output":131072},"cost":{"input":0.05,"output":0.2,"cache_read":0.005},"knowledge":"2026-02","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "upstage/solar-pro4": {"name":"Solar Pro 4","description":"Flagship model for demanding analysis, coding, and production agent workflows","family":"solar-pro","canonical_model_id":"upstage/solar-pro4","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":524288,"output":131072},"cost":{"input":0.15,"output":0.6,"cache_read":0.03},"knowledge":"2026-02","release_date":"2026-08-06","last_updated":"2026-08-06","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "writer/palmyra-x5": {"name":"Palmyra X5","description":"General-purpose chat model for instruction following, writing, and analysis","family":"palmyra","canonical_model_id":"writer/palmyra-x5","reasoning":false,"tool_call":false,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1040000,"output":8192},"cost":{"input":0.6,"output":6},"release_date":"2025-04-28","last_updated":"2025-04-28","open_weights":false},
  "x-ai/grok-4.20": {"name":"Grok 4.20","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":2000000,"output":1800000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"knowledge":"2025-09-01","release_date":"2026-03-31","last_updated":"2026-03-31","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "x-ai/grok-4.20-multi-agent": {"name":"Grok 4.20 Multi-Agent","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":2000000,"output":1800000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"knowledge":"2025-09-01","release_date":"2026-03-31","last_updated":"2026-03-31","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "x-ai/grok-4.3": {"name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":900000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-04-17","last_updated":"2026-04-17","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "x-ai/grok-4.5": {"name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":450000},"cost":{"input":2,"output":6,"cache_read":0.3,"tiers":[{"input":4,"output":12,"cache_read":0.6,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":0.6}},"release_date":"2026-07-08","last_updated":"2026-07-08","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "x-ai/grok-4.6": {"name":"Grok 4.6","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":450000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "x-ai/grok-4.7": {"name":"Grok 4.7","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":450000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "x-ai/grok-build-0.1": {"name":"Grok Build 0.1","description":"Fast Grok coding model tuned for agentic engineering and iterative edits","family":"grok-build","canonical_model_id":"xai/grok-build-0.1","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":256000,"output":230400},"cost":{"input":1,"output":2,"cache_read":0.2,"tiers":[{"input":2,"output":4,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2,"output":4,"cache_read":0.4}},"release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false},
  "xiaomi/mimo-v2.5": {"name":"MiMo-V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1050000,"output":65536},"cost":{"input":0.4,"output":2,"cache_read":0.08},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "xiaomi/mimo-v2.5-pro": {"name":"MiMo-V2.5-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1050000,"output":235929},"cost":{"input":0.48,"output":1.8,"cache_read":0.096},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "xiaomi/mimo-v2.6-flash": {"name":"MiMo-V2.6-Flash","description":"MiMo flash model for fast multimodal assistance and agent workflows","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1050000,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "xiaomi/mimo-v2.6-pro": {"name":"MiMo-V2.6-Pro","description":"MiMo pro model for strong multimodal reasoning and agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1050000,"output":131072},"cost":{"input":0.435,"output":0.87,"cache_read":0.0036},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "xiaomi/mimo-v2.6-pro-ultraspeed": {"name":"MiMo-V2.6-Pro-UltraSpeed","description":"MiMo pro model for strong multimodal reasoning and agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-pro-ultraspeed","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":4.35,"output":8.7,"cache_read":0.036},"release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "xingchen-agi/xing4.0-29b-a4b:free": {"name":"Xing4.0 29B A4B (free)","description":"Free provider route for experiments, demos, and cost-sensitive chat workloads","family":"agi","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":235929},"cost":{"input":0,"output":0},"release_date":"2026-10-16","last_updated":"2026-10-16","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-4.5": {"name":"GLM-4.5","description":"Hybrid-reasoning GLM release that made the 4.5 line broadly useful","family":"glm","canonical_model_id":"zhipuai/glm-4.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":98304},"cost":{"input":0.6,"output":2.2,"cache_read":0.11},"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":true,"status":"deprecated","rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-4.5-air": {"name":"GLM-4.5-Air","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","canonical_model_id":"zhipuai/glm-4.5-air","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":98304},"cost":{"input":0.13,"output":0.85,"cache_read":0.025},"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-4.5v": {"name":"GLM-4.5V","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","canonical_model_id":"zhipuai/glm-4.5v","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":65536,"output":16384},"cost":{"input":0.6,"output":1.8,"cache_read":0.11},"knowledge":"2025-04","release_date":"2025-08-11","last_updated":"2025-08-11","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-4.6": {"name":"GLM-4.6","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.5,"output":2,"cache_read":0.1},"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-4.6v": {"name":"GLM-4.6V","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","canonical_model_id":"zhipuai/glm-4.6v","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.3,"output":0.9,"cache_read":0.055},"knowledge":"2025-04","release_date":"2025-12-08","last_updated":"2025-12-08","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-4.7": {"name":"GLM-4.7","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.54,"output":1.98,"cache_read":0.099},"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-4.7-flash": {"name":"GLM-4.7-Flash","description":"Budget GLM lane for fast coding help, routing, and everyday automation","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_details"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":117964},"cost":{"input":0.0605,"output":0.4},"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5": {"name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5-turbo": {"name":"GLM-5-Turbo","description":"Faster GLM-5 lane for coding agents that need lower latency","family":"glm","canonical_model_id":"zhipuai/glm-5-turbo","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24},"release_date":"2026-03-16","last_updated":"2026-03-16","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5.1": {"name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","canonical_model_id":"zhipuai/glm-5.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-04-07","last_updated":"2026-04-07","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5.2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.006,"output":7,"cache_read":0.005},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5.3": {"name":"GLM-5.3","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.08,"output":3.6,"cache_read":0.076},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5.3-flash": {"name":"GLM-5.3-Flash","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.0197,"output":0.1324,"cache_read":0.0197},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5.3-flashx": {"name":"GLM 5.3 FlashX","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.37,"output":1.25,"cache_read":0.09},"release_date":"2026-09-18","last_updated":"2026-09-18","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5.3-prime": {"name":"GLM 5.3 Prime","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":2.8,"output":8.8,"cache_read":0.56},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "z-ai/glm-5v-turbo": {"name":"GLM-5V-Turbo","description":"Fast GLM vision model for screenshots, documents, and multimodal agent tasks","family":"glm","canonical_model_id":"zhipuai/glm-5v-turbo","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["image","text","video"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24},"release_date":"2026-04-01","last_updated":"2026-04-01","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~anthropic/claude-fable-latest": {"name":"Claude Fable Latest","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~anthropic/claude-haiku-latest": {"name":"Claude Haiku Latest","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~anthropic/claude-opus-latest": {"name":"Claude Opus Latest","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"release_date":"2026-04-21","last_updated":"2026-04-21","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~anthropic/claude-sonnet-latest": {"name":"Claude Sonnet Latest","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~deepseek/deepseek-flash-latest": {"name":"DeepSeek Flash Latest","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.022,"output":0.192,"cache_read":0.004},"release_date":"2026-09-14","last_updated":"2026-09-14","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~deepseek/deepseek-pro-latest": {"name":"DeepSeek Pro Latest","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":393216},"cost":{"input":0.1035,"output":4,"cache_read":0.1},"release_date":"2026-09-14","last_updated":"2026-09-14","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~deepseek/deepseek-v4-flash-latest": {"name":"DeepSeek V4 Flash Latest","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.0429,"output":0.1677,"cache_read":0.028},"release_date":"2026-08-01","last_updated":"2026-08-01","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~google/gemini-flash-latest": {"name":"Gemini Flash Latest","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf","audio"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"reasoning":3.75,"cache_read":0.075,"cache_write":0.041667,"input_audio":0.75},"knowledge":"2025-01-01","release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~google/gemini-pro-latest": {"name":"Gemini Pro Latest","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["audio","pdf","image","text","video"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":2,"output":10,"reasoning":10,"cache_read":0.1,"cache_write":0.1875,"input_audio":2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~moonshotai/kimi-latest": {"name":"Kimi Latest","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.1634,"output":10.5539,"cache_read":0.16},"release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~openai/gpt-astra-latest": {"name":"GPT Astra Latest","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"release_date":"2026-09-11","last_updated":"2026-09-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~openai/gpt-luna-latest": {"name":"GPT Luna Latest","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-02-16","release_date":"2026-09-11","last_updated":"2026-09-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~openai/gpt-mini-latest": {"name":"GPT Mini Latest","description":"Compact GPT model for low-latency assistance and high-volume workloads","family":"gpt-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":400000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075},"knowledge":"2025-08-31","release_date":"2026-04-27","last_updated":"2026-04-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~openai/gpt-sol-latest": {"name":"GPT Sol Latest","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-09-11","last_updated":"2026-09-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~openai/gpt-terra-latest": {"name":"GPT Terra Latest","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","none","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["pdf","image","text"],"output":["text"]},"limit":{"context":1050000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-02-16","release_date":"2026-09-11","last_updated":"2026-09-11","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~unbiased/pareto-latest": {"name":"Pareto Latest","description":"Multimodal model for analyzing text, images, documents, and rich media","reasoning":false,"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.8,"output":3.2,"cache_read":0.03},"release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "~x-ai/grok-latest": {"name":"Grok Latest","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":450000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"release_date":"2026-07-08","last_updated":"2026-07-08","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~z-ai/glm-flash-latest": {"name":"GLM Flash Latest","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.0197,"output":0.1324,"cache_read":0.0197},"release_date":"2026-08-27","last_updated":"2026-08-27","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "~z-ai/glm-latest": {"name":"GLM Latest","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]},{"type":"budget_tokens"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":943718},"cost":{"input":0.0699,"output":1.4,"cache_read":0.0689},"release_date":"2026-08-19","last_updated":"2026-08-19","open_weights":false,"rig":{"pinned":["reasoning_options"]}}
}},
"perplexity": {"models": {
  "sonar": {"name":"Sonar","description":"Fast web-grounded Sonar for current answers, citations, and lightweight retrieval","family":"sonar","canonical_model_id":"perplexity/sonar","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":1,"output":1},"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","open_weights":false,"status":"deprecated"},
  "sonar-deep-research": {"name":"Perplexity Sonar Deep Research","description":"Sonar search model for current answers, retrieval, and citation-backed chat","family":"sonar","canonical_model_id":"perplexity/sonar-deep-research","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":2,"output":8,"reasoning":3},"knowledge":"2025-01","release_date":"2025-02-01","last_updated":"2025-09-01","open_weights":false,"status":"deprecated"},
  "sonar-pro": {"name":"Sonar Pro","description":"Deeper Sonar search model with broader retrieval and stronger synthesis","family":"sonar-pro","canonical_model_id":"perplexity/sonar-pro","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":8192},"cost":{"input":3,"output":15},"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","open_weights":false,"status":"deprecated"},
  "sonar-reasoning-pro": {"name":"Sonar Reasoning Pro","description":"Web-grounded Sonar for multi-step research questions that need cited reasoning","family":"sonar-reasoning","canonical_model_id":"perplexity/sonar-reasoning-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":2,"output":8},"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","open_weights":false,"status":"deprecated"},
  "sonar_pro": {"name":"Sonar Pro","description":"Deeper Sonar search model with broader retrieval and stronger synthesis","family":"sonar-pro","canonical_model_id":"perplexity/sonar-pro","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":8192},"cost":{"input":3,"output":15},"knowledge":"2025-09-01","release_date":"2024-01-01","last_updated":"2025-09-01","open_weights":false,"status":"deprecated","rig":{"pinned":["name","description","family","reasoning","tool_call","temperature","attachment","modalities","limit.context","limit.output","cost.input","cost.output","knowledge","release_date","last_updated","open_weights","status"]}}
}},
"together": {"models": {
  "BAAI/bge-base-en-v1.5": {"name":"BAAI/bge-base-en-v1.5","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "BAAI/bge-large-en-v1.5": {"name":"BAAI/bge-large-en-v1.5","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "LiquidAI/LFM2-24B-A2B": {"name":"LFM2-24B-A2B","description":"Efficient open Liquid MoE instruct model for on-device chat and tool calling","family":"liquid","canonical_model_id":"liquid/lfm2-24b-a2b","reasoning":false,"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":32768},"cost":{"input":0.03,"output":0.12},"release_date":"2026-02-24","last_updated":"2026-02-24","open_weights":true,"status":"deprecated"},
  "MiniMaxAI/MiniMax-M2.5": {"name":"MiniMax-M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.5","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"status":"deprecated"},
  "MiniMaxAI/MiniMax-M2.7": {"name":"MiniMax-M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.7","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":196608,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true,"status":"deprecated"},
  "MiniMaxAI/MiniMax-M3": {"name":"MiniMax-M3","description":"MiniMax multimodal model for long-context coding, perception, and agent planning","family":"minimax","canonical_model_id":"minimax/MiniMax-M3","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":524288,"output":524288},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-06-01","last_updated":"2026-06-01","open_weights":true},
  "Qwen/Qwen2.5-72B-Instruct-Turbo": {"name":"Qwen2.5 72B Instruct Turbo","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "Qwen/Qwen2.5-7B-Instruct-Turbo": {"name":"Qwen 2.5 7B Instruct Turbo","description":"Small open Qwen2.5 instruction model for chat, structured output, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen2.5-7b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":32768},"cost":{"input":0.3,"output":0.3},"release_date":"2024-09-19","last_updated":"2024-09-19","open_weights":true,"status":"deprecated"},
  "Qwen/Qwen3-235B-A22B-Instruct-2507-tput": {"name":"Qwen3 235B A22B Instruct 2507 FP8","description":"Updated large open Qwen3 MoE instruct model for multilingual chat, coding, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-instruct-2507","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.2,"output":0.6},"release_date":"2025-07-21","last_updated":"2025-07-21","open_weights":true,"status":"deprecated"},
  "Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8": {"name":"Qwen3 Coder 480B A35B Instruct","description":"Open Qwen coding heavyweight for repository reasoning and agentic engineering","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-480b-a35b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":2,"output":2},"knowledge":"2025-04","release_date":"2025-07-22","last_updated":"2025-07-22","open_weights":true,"status":"deprecated"},
  "Qwen/Qwen3-Coder-Next-FP8": {"name":"Qwen3 Coder Next FP8","description":"Open-weight Qwen coding model for agents, repository edits, and multi-turn tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-coder-next","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.5,"output":1.2},"knowledge":"2025-09","release_date":"2026-02-03","last_updated":"2026-02-03","open_weights":true,"status":"deprecated"},
  "Qwen/Qwen3.5-397B-A17B": {"name":"Qwen3.5 397B A17B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-397b-a17b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":130000},"cost":{"input":0.6,"output":3.6,"cache_read":0.35},"release_date":"2026-02-15","last_updated":"2026-02-15","open_weights":true,"status":"deprecated"},
  "Qwen/Qwen3.5-9B": {"name":"Qwen3.5 9B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.5-9b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.17,"output":0.25},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true},
  "Qwen/Qwen3.6-Plus": {"name":"Qwen3.6 Plus","description":"Earlier Qwen multimodal workhorse for million-token agent and document tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.6-plus","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.5,"output":3},"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":false},
  "Qwen/Qwen3.7-Max": {"name":"Qwen3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.7-max","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":2.5,"output":7.5,"cache_read":0.5},"release_date":"2026-05-21","last_updated":"2026-05-21","open_weights":false},
  "WhereIsAI/UAE-Large-V1": {"name":"WhereIsAI/UAE-Large-V1","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "bert-base-uncased": {"name":"bert-base-uncased","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "deepseek-ai/DeepSeek-R1": {"name":"DeepSeek-R1","description":"Classic open reasoning model for transparent math, coding, and deliberate problem solving","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-r1","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163839,"output":163839},"cost":{"input":3,"output":7},"knowledge":"2024-07","release_date":"2025-01-20","last_updated":"2025-05-29","open_weights":true,"status":"deprecated"},
  "deepseek-ai/DeepSeek-V3": {"name":"DeepSeek-V3","description":"Open DeepSeek MoE chat model for coding, math, and general reasoning","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":1.25,"output":1.25},"release_date":"2024-12-26","last_updated":"2024-12-26","open_weights":true,"status":"deprecated"},
  "deepseek-ai/DeepSeek-V3-1": {"name":"DeepSeek V3.1","description":"Hybrid-reasoning DeepSeek model with thinking and non-thinking modes","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3.1","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.6,"output":1.7},"release_date":"2025-08-21","last_updated":"2025-08-21","open_weights":true,"status":"deprecated"},
  "deepseek-ai/DeepSeek-V4-Flash-0731": {"name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-0731","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.14,"output":0.28,"cache_read":0.03},"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","open_weights":true,"status":"deprecated"},
  "deepseek-ai/DeepSeek-V4-Pro-0813": {"name":"DeepSeek V4 Pro 0813","description":"DeepSeek V4 Pro snapshot with million-token context and support for thinking and non-thinking modes","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro-0813","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":1.32,"output":3.96,"cache_read":0.13},"release_date":"2026-08-13","last_updated":"2026-08-22","open_weights":true},
  "deepseek-ai/DeepSeek-V4.1-Flash": {"name":"DeepSeek V4.1 Flash","description":"DeepSeek V4.1 Flash model for reasoning and agentic coding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":384000},"cost":{"input":0.3,"output":1.2,"cache_read":0.006},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true},
  "essentialai/Rnj-1-Instruct": {"name":"Rnj-1 Instruct","description":"Open 8B Essential AI model for code, STEM reasoning, and agentic tool use","family":"rnj","canonical_model_id":"essentialai/rnj-1-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":32768,"output":32768},"cost":{"input":0.15,"output":0.15},"release_date":"2025-12-04","last_updated":"2025-12-04","open_weights":true,"status":"deprecated"},
  "hazyresearch/M2-BERT-2k-Retrieval-Encoder-V1": {"name":"hazyresearch/M2-BERT-2k-Retrieval-Encoder-V1","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "meta-llama/Llama-3-8b-chat-hf": {"name":"Llama 3 8B Chat","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Llama-3.3-70B-Instruct-Turbo": {"name":"Llama 3.3 70B","description":"Popular open Llama workhorse for multilingual chat, coding, and self-hosting","family":"llama","canonical_model_id":"meta/llama-3.3-70b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":1.04,"output":1.04},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true,"status":"deprecated"},
  "meta-llama/Meta-Llama-3-70B-Instruct": {"name":"Llama 3 70B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Meta-Llama-3-8B-Instruct": {"name":"Llama 3 8B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo": {"name":"Llama 3.1 70B Instruct Turbo","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo": {"name":"Llama 3.1 8B Instruct Turbo","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "meta-models/Muse-Spark-Open-816B-A42B": {"name":"Muse Spark Open","description":"Muse Spark Open is an 816B total / 42B active open-weight multimodal reasoning model from Meta for coding and long-horizon agentic work, with 1M-token context.","family":"muse","canonical_model_id":"meta/muse-spark-open","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.25,"output":4.25,"cache_read":0.15},"knowledge":"2026-01-04","release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true},
  "mistralai/Mixtral-8x7B-Instruct-v0.1": {"name":"Mixtral 8x7B Instruct","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "moonshotai/Kimi-K3": {"name":"Kimi K3","description":"Multimodal Kimi model with 1M context and toggleable max-effort thinking for long-horizon agent work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.3},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "nvidia/NVIDIA-Nemotron-3.5-Super-VL": {"name":"Nemotron 3.5 Super VL 120B A12B","description":"NVIDIA vision-language reasoning model for text, image, and video understanding and agentic tool use","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-super-vl-120b-a12b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.15,"output":0.5,"cache_read":0.02},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true},
  "nvidia/nemotron-3-ultra-550b-a55b": {"name":"Nemotron 3 Ultra 550B A55B","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-ultra-550b-a55b","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":512300,"output":512300},"cost":{"input":0.6,"output":3.6,"cache_read":0.2},"knowledge":"2025-09","release_date":"2026-06-04","last_updated":"2026-06-04","open_weights":true,"status":"deprecated"},
  "openai/gpt-oss-120b": {"name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.15,"output":0.6},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "reflection/Beam-1.0-501B-A23B": {"name":"Beam 1.0 501B A23B","description":"Sparse mixture-of-experts reasoning model with 23B active parameters for coding and agentic workflows","family":"beam","canonical_model_id":"reflection/beam-1.0-501b-a23b","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.4,"output":1.2,"cache_read":0.04},"knowledge":"2026-06-30","release_date":"2026-10-05","last_updated":"2026-10-16","open_weights":true},
  "sentence-transformers/msmarco-bert-base-dot-v5": {"name":"sentence-transformers/msmarco-bert-base-dot-v5","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "thinkingmachines/Inkling": {"name":"Inkling","description":"Multimodal MoE reasoning model (975B total, 41B active) for text, image, and audio","family":"ling","canonical_model_id":"thinkingmachines/inkling","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","minimal","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio"],"output":["text"]},"limit":{"context":524288,"output":131072},"cost":{"input":1,"output":4.05,"cache_read":0.17},"release_date":"2026-07-15","last_updated":"2026-07-15","open_weights":true},
  "togethercomputer/llama-2-70b-chat": {"name":"Llama 2 70B Chat","reasoning":false,"tool_call":true,"temperature":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","tool_call","temperature","modalities"]}},
  "togethercomputer/m2-bert-80M-2k-retrieval": {"name":"togethercomputer/m2-bert-80M-2k-retrieval","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "togethercomputer/m2-bert-80M-32k-retrieval": {"name":"togethercomputer/m2-bert-80M-32k-retrieval","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "togethercomputer/m2-bert-80M-8k-retrieval": {"name":"togethercomputer/m2-bert-80M-8k-retrieval","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "zai-org/GLM-5": {"name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":1,"output":3.2},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"status":"deprecated"},
  "zai-org/GLM-5.1": {"name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","canonical_model_id":"zhipuai/glm-5.1","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":202752,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-04-07","last_updated":"2026-04-07","open_weights":true,"status":"deprecated"},
  "zai-org/GLM-5.2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048575,"output":164000},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true},
  "zai-org/GLM-5.3": {"name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048575,"output":262144},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "zai-org/GLM-5.3-Flash": {"name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048575,"output":400000},"cost":{"input":0.15,"output":0.5,"cache_read":0.03},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true}
}},
"venice": {"models": {
  "abliteration-abliterated-model-large-flash": {"name":"Abliterated Large Flash","description":"Multimodal reasoning model for visual analysis, planning, and tool use","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.25,"output":2.5,"cache_read":0.125},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":false},
  "abliteration-abliterated-model-large-v2": {"name":"Abliterated Large V2","description":"Flagship model for demanding analysis, coding, and production agent workflows","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":3,"output":5,"cache_read":0.3},"release_date":"2026-10-01","last_updated":"2026-10-01","open_weights":false},
  "aion-labs-aion-3-0": {"name":"Aion 3.0","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","canonical_model_id":"aion-labs/aion-3.0","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":3.75,"output":7.5,"cache_read":0.9375},"release_date":"2026-05-05","last_updated":"2026-05-05","open_weights":false},
  "aion-labs-aion-3-0-mini": {"name":"Aion 3.0 Mini","description":"Efficient model for low-latency assistance, extraction, and routine automation","canonical_model_id":"aion-labs/aion-3.0-mini","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":0.875,"output":1.75,"cache_read":0.225},"release_date":"2026-05-14","last_updated":"2026-05-14","open_weights":false},
  "aion-labs-aion-3-5": {"name":"Aion 3.5","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":3.75,"output":7.5,"cache_read":0.9375},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false},
  "aion-labs-aion-3-5-mini": {"name":"Aion 3.5 Mini","description":"Efficient model for low-latency assistance, extraction, and routine automation","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.875,"output":1.75,"cache_read":0.225},"release_date":"2026-09-23","last_updated":"2026-09-23","open_weights":false},
  "claude-fable-5": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false},
  "claude-fable-5-1": {"name":"Claude Fable 5.1","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false},
  "claude-haiku-5-5": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false},
  "claude-opus-4-5": {"name":"Claude Opus 4.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":198000,"output":32768},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false},
  "claude-opus-4-6": {"name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false},
  "claude-opus-4-7": {"name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false},
  "claude-opus-4-8": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "claude-opus-4-8-fast": {"name":"Claude Opus 4.8 Fast","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "claude-opus-5": {"name":"Claude Opus 5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false},
  "claude-opus-5-5": {"name":"Claude Opus 5.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "claude-opus-5-5-fast": {"name":"Claude Opus 5.5 Fast","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":8,"output":40,"cache_read":0.4,"cache_write":10},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "claude-opus-5-fast": {"name":"Claude Opus 5 Fast","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false},
  "claude-sonnet-4-6": {"name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false},
  "claude-sonnet-5": {"name":"Claude Sonnet 5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false},
  "claude-sonnet-5-5": {"name":"Claude Sonnet 5.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false},
  "deepseek-v3.2": {"name":"DeepSeek V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","canonical_model_id":"deepseek/deepseek-v3.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":160000,"output":32768},"cost":{"input":0.33,"output":0.48,"cache_read":0.16},"knowledge":"2024-07","release_date":"2025-12-01","last_updated":"2025-12-01","open_weights":true,"status":"deprecated"},
  "deepseek-v4-1-flash": {"name":"DeepSeek V4.1 Flash","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.0075},"knowledge":"2025-05","release_date":"2026-09-10","last_updated":"2026-09-10","open_weights":true},
  "deepseek-v4-1-flash-1017": {"name":"DeepSeek v4.1 Flash (1017)","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4.1-flash-1017","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.3,"output":1.2,"cache_read":0.0075},"knowledge":"2025-05","release_date":"2026-10-17","last_updated":"2026-10-17","open_weights":true},
  "deepseek-v4-1-flash-1017-uncensored": {"name":"DeepSeek V4.1 Flash 1017 Uncensored","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.33,"output":1.32,"cache_read":0.00825},"release_date":"2026-10-27","last_updated":"2026-10-27","open_weights":true},
  "deepseek-v4-1-flash-uncensored": {"name":"DeepSeek V4.1 Flash Uncensored","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.33,"output":1.32,"cache_read":0.00825},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":true},
  "deepseek-v4-flash": {"name":"DeepSeek V4 Flash 0423","description":"Fast DeepSeek V4 lane for economical reasoning, coding, and long-context work","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":0.138,"output":0.275,"cache_read":0.028},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true},
  "deepseek-v4-flash-0731": {"name":"DeepSeek V4 Flash 0731","description":"Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-0731","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":0.13,"output":0.26,"cache_read":0.028},"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","open_weights":true},
  "deepseek-v4-flash-0731-fast": {"name":"DeepSeek V4 Flash 0731 Fast","description":"Fast DeepSeek model for efficient chat, coding help, and agent loops","family":"deepseek-flash","canonical_model_id":"deepseek/deepseek-v4-flash-0731","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":0.35,"output":0.7,"cache_read":0.0875},"knowledge":"2025-05","release_date":"2026-07-31","last_updated":"2026-07-31","open_weights":true},
  "deepseek-v4-pro": {"name":"DeepSeek V4 Pro","description":"Open MoE flagship with million-token context for coding and long agent runs","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":1.65,"output":3.301,"cache_read":0.33},"knowledge":"2025-05","release_date":"2026-04-24","last_updated":"2026-04-24","open_weights":true},
  "deepseek-v4-pro-0813": {"name":"DeepSeek V4 Pro 0813","description":"Flagship DeepSeek model for coding, reasoning, and agentic work","family":"deepseek-thinking","canonical_model_id":"deepseek/deepseek-v4-pro-0813","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":1.65,"output":4.95,"cache_read":0.165},"release_date":"2026-08-13","last_updated":"2026-08-22","open_weights":true},
  "elevenlabs/scribe-v2": {"name":"elevenlabs/scribe-v2","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "fal-ai/wizper": {"name":"fal-ai/wizper","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "flux-2-pro": {"name":"flux-2-pro","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "gemini-3-1-pro-preview": {"name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","canonical_model_id":"google/gemini-3.1-pro-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":32768},"cost":{"input":2.5,"output":15,"cache_read":0.5,"cache_write":0.5,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"cache_write":0.5,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5,"cache_write":0.5}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false},
  "gemini-3-5-flash": {"name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.5-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":1.55,"output":9.45,"cache_read":0.155,"cache_write":0.086},"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","open_weights":false},
  "gemini-3-5-flash-lite": {"name":"Gemini 3.5 Flash-Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.5-flash-lite","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.375,"output":3.125,"cache_read":0.0375},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false},
  "gemini-3-6-flash": {"name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.6-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.9375,"output":4.6875,"cache_read":0.09375},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false},
  "gemini-3-7-flash": {"name":"Gemini 3.7 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.7-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.9375,"output":4.6875,"cache_read":0.09375},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false},
  "gemini-3-8-flash": {"name":"Gemini 3.8 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.8-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.9375,"output":4.6875,"cache_read":0.09375},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false},
  "gemini-3-flash-preview": {"name":"Gemini 3 Flash Preview","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","canonical_model_id":"google/gemini-3-flash-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.7,"output":3.75,"cache_read":0.07},"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","open_weights":false},
  "gemini-4-argon": {"name":"Gemini 4 Argon","description":"Gemini multimodal model for text, image, audio, video, and document tasks","family":"gemini","canonical_model_id":"google/gemini-4-argon","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":2.5,"output":12.5,"cache_read":0.125},"knowledge":"2026-03","release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "gemma-4-uncensored": {"name":"Gemma 4 Uncensored","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","reasoning":false,"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":8192},"cost":{"input":0.1625,"output":0.5},"release_date":"2026-04-13","last_updated":"2026-06-11","open_weights":true},
  "google-gemma-3-27b-it": {"name":"Google Gemma 3 27B Instruct","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-3-27b-it","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":198000,"output":16384},"cost":{"input":0.12,"output":0.2},"knowledge":"2024-08","release_date":"2025-03-12","last_updated":"2025-03-12","open_weights":true},
  "google-gemma-4-26b-a4b-it": {"name":"Google Gemma 4 26B A4B Instruct","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-26b-a4b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":256000,"output":8192},"cost":{"input":0.13,"output":0.4,"cache_read":0.05},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true},
  "google-gemma-4-31b-it": {"name":"Google Gemma 4 31B Instruct","description":"Largest Gemma 4 instruction model for open, self-hosted chat and reasoning","family":"gemma","canonical_model_id":"google/gemma-4-31b-it","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":256000,"output":8192},"cost":{"input":0.12,"output":0.36,"cache_read":0.09},"knowledge":"2025-01","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true},
  "grok-4-20": {"name":"Grok 4.20","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":2000000,"output":128000},"cost":{"input":1.42,"output":2.83,"cache_read":0.23,"tiers":[{"input":2.83,"output":5.67,"cache_read":0.45,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.83,"output":5.67,"cache_read":0.45}},"release_date":"2026-03-12","last_updated":"2026-06-11","open_weights":false},
  "grok-4-20-multi-agent": {"name":"Grok 4.20 Multi-Agent","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":true,"reasoning_options":[],"tool_call":false,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":2000000,"output":128000},"cost":{"input":1.42,"output":2.83,"cache_read":0.23,"tiers":[{"input":2.83,"output":5.67,"cache_read":0.45,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.83,"output":5.67,"cache_read":0.45}},"release_date":"2026-03-12","last_updated":"2026-06-11","open_weights":false},
  "grok-4-3": {"name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":32000},"cost":{"input":1.42,"output":2.83,"cache_read":0.23,"tiers":[{"input":2.83,"output":5.67,"cache_read":0.45,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.83,"output":5.67,"cache_read":0.45}},"release_date":"2026-04-17","last_updated":"2026-04-17","open_weights":false},
  "grok-4-5": {"name":"Grok 4.5","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","canonical_model_id":"xai/grok-4.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":32000},"cost":{"input":2.27,"output":6.8,"cache_read":0.34,"tiers":[{"input":4.53,"output":13.6,"cache_read":0.68,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4.53,"output":13.6,"cache_read":0.68}},"release_date":"2026-07-08","last_updated":"2026-07-08","open_weights":false},
  "grok-4-6": {"name":"Grok 4.6","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":200000},"cost":{"input":2.27,"output":6.8,"cache_read":0.57,"tiers":[{"input":4.53,"output":13.6,"cache_read":1.13,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4.53,"output":13.6,"cache_read":1.13}},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":false},
  "grok-4-7": {"name":"Grok 4.7","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":500000,"output":200000},"cost":{"input":2.27,"output":6.8,"cache_read":0.57,"tiers":[{"input":4.53,"output":13.6,"cache_read":1.13,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4.53,"output":13.6,"cache_read":1.13}},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false},
  "grok-build-0-1": {"name":"Grok Build 0.1","description":"Fast Grok coding model tuned for agentic engineering and iterative edits","family":"grok-build","canonical_model_id":"xai/grok-build-0.1","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":1,"output":2,"cache_read":0.2,"tiers":[{"input":2,"output":4,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2,"output":4,"cache_read":0.4}},"release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false},
  "hermes-3-llama-3.1-405b": {"name":"Hermes 3 Llama 3.1 405b","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"nousresearch","canonical_model_id":"nousresearch/hermes-3-llama-3.1-405b","reasoning":false,"tool_call":false,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":1.1,"output":3},"knowledge":"2023-12-31","release_date":"2024-08-16","last_updated":"2024-08-16","open_weights":true,"status":"deprecated"},
  "hunyuan-image-v3": {"name":"hunyuan-image-v3","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "hy4-preview": {"name":"Hy4 Preview","description":"Preview model for early access evaluation, prototyping, and compatibility testing","family":"Hy","canonical_model_id":"tencent/hy4-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":64000},"cost":{"input":1.0425,"output":3.12625,"cache_read":0.0525},"release_date":"2026-08-28","last_updated":"2026-08-28","open_weights":true},
  "kimi-k2-5": {"name":"Kimi K2.5","description":"Earlier Kimi frontier model for long-context agents, coding, and multimodal work","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.56,"output":3.5,"cache_read":0.22},"knowledge":"2025-01","release_date":"2026-01-27","last_updated":"2026-01-27","open_weights":true},
  "kimi-k2-6": {"name":"Kimi K2.6","description":"Multimodal Kimi workhorse for agent loops, coding tasks, and visual context","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.75,"output":3.5,"cache_read":0.16},"knowledge":"2025-01","release_date":"2026-04-20","last_updated":"2026-04-20","open_weights":true},
  "kimi-k2-7-code": {"name":"Kimi K2.7 Code","description":"Coding-focused Kimi model, stronger on long-horizon repo work with less overthinking","family":"kimi-k2","canonical_model_id":"moonshotai/kimi-k2.7-code","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.75,"output":3.5,"cache_read":0.16},"knowledge":"2025-01","release_date":"2026-06-12","last_updated":"2026-06-12","open_weights":true},
  "kimi-k3": {"name":"Kimi K3","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":3.75,"output":18.75,"cache_read":0.375},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "kimi-k3-1-pro": {"name":"Kimi K3.1 Pro","description":"Multimodal Kimi model with 1M context and always-on reasoning for coding, 3D creation, financial research, and agentic work","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3.1-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":3,"output":15,"cache_read":0.30000000000000004},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":false},
  "kimi-k3-1-pro-highspeed": {"name":"Kimi K3.1 Pro Highspeed","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":6,"output":30,"cache_read":0.6000000000000001},"release_date":"2026-10-23","last_updated":"2026-10-23","open_weights":false},
  "kimi-k3-fast-api": {"name":"Kimi K3 Fast","description":"Kimi multimodal agent model for visual understanding, coding, and planning","family":"kimi-k3","canonical_model_id":"moonshotai/kimi-k3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":4.5,"output":22.5,"cache_read":0.45},"release_date":"2026-07-16","last_updated":"2026-07-16","open_weights":true},
  "llama-3.2-3b": {"name":"Llama 3.2 3B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.2-3b","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.15,"output":0.6},"knowledge":"2023-12","release_date":"2024-09-25","last_updated":"2024-09-25","open_weights":true},
  "llama-3.3-70b": {"name":"Llama 3.3 70B","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","canonical_model_id":"meta/llama-3.3-70b-instruct","reasoning":false,"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.7,"output":2.8},"knowledge":"2023-12","release_date":"2024-12-06","last_updated":"2024-12-06","open_weights":true},
  "mercury-2": {"name":"Mercury 2","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"mercury","canonical_model_id":"inception/mercury-2","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":50000},"cost":{"input":0.3125,"output":0.9375,"cache_read":0.03125},"release_date":"2026-02-24","last_updated":"2026-02-24","open_weights":false},
  "mercury-2-5": {"name":"Mercury 2.5","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"mercury","canonical_model_id":"inception/mercury-2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":260000,"output":65536},"cost":{"input":0.25,"output":0.9375,"cache_read":0.025},"release_date":"2026-09-08","last_updated":"2026-09-08","open_weights":false},
  "mercury-3": {"name":"Mercury 3","description":"Reasoning model for deliberate analysis, multi-step problem solving, and tool use","family":"mercury","canonical_model_id":"inception/mercury-3","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.1,"output":0.35,"cache_read":0.01},"release_date":"2026-10-16","last_updated":"2026-10-20","open_weights":false},
  "minimax-m25": {"name":"MiniMax M2.5","description":"Prior MiniMax coding model for agent workflows, office edits, and automation","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":198000,"output":32768},"cost":{"input":0.27,"output":0.95,"cache_read":0.03},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true},
  "minimax-m27": {"name":"MiniMax M2.7","description":"Open MiniMax flagship for coding agents, office automation, and complex environments","family":"minimax","canonical_model_id":"minimax/MiniMax-M2.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":198000,"output":32768},"cost":{"input":0.375,"output":1.5,"cache_read":0.06875},"release_date":"2026-03-18","last_updated":"2026-03-18","open_weights":true},
  "minimax-m3-1-flash": {"name":"MiniMax M3.1 Flash","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax","canonical_model_id":"minimax/MiniMax-M3.1-Flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.1875,"output":0.75,"cache_read":0.01875},"release_date":"2026-10-19","last_updated":"2026-10-19","open_weights":false,"rig":{"pinned":["reasoning_options"]}},
  "minimax-m3-preview": {"name":"MiniMax M3 Preview","description":"MiniMax multimodal coding model for long-context reasoning and agent tasks","family":"minimax-m3","reasoning":true,"reasoning_options":[],"tool_call":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":524288,"output":65536},"cost":{"input":0.3,"output":1.2,"cache_read":0.06},"release_date":"2026-06-12","last_updated":"2026-06-13","open_weights":true},
  "mistral-small-2603": {"name":"Mistral Small (26.03)","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"modalities":{"input":["text","image"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","temperature","modalities"]}},
  "mistral-small-3-2-24b-instruct": {"name":"Mistral Small 3.2 24B Instruct","description":"Efficient Mistral model for fast chat, extraction, and production assistants","family":"mistral-small","canonical_model_id":"mistral/mistral-small-3.2-24b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["image","text"],"output":["text"]},"limit":{"context":256000,"output":16384},"cost":{"input":0.09375,"output":0.25},"knowledge":"2023-10-31","release_date":"2025-06-20","last_updated":"2025-06-20","open_weights":true},
  "muse-spark-open": {"name":"Muse Spark Open","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"muse","canonical_model_id":"meta/muse-spark-open","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.5625,"output":5.3125,"cache_read":0.1875},"knowledge":"2026-01-04","release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true},
  "muse-wonder": {"name":"Muse Wonder","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"muse","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","max"]}],"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":5,"output":25,"cache_read":0.25},"release_date":"2026-10-12","last_updated":"2026-10-15","open_weights":false},
  "nvidia-nemotron-3-5-super-vl-120b-a12b": {"name":"NVIDIA Nemotron 3.5 Super VL 120B","description":"Nemotron multimodal model for visual reasoning and agentic AI workflows","family":"nemotron","canonical_model_id":"nvidia/nemotron-3.5-super-vl-120b-a12b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":32768},"cost":{"input":0.2,"output":0.5,"cache_read":0.04},"release_date":"2026-10-15","last_updated":"2026-10-15","open_weights":true},
  "nvidia-nemotron-3-nano-30b-a3b": {"name":"NVIDIA Nemotron 3 Nano 30B","description":"Small Nemotron 3 MoE for efficient coding, math, and long-context agents","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-nano-30b-a3b","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.075,"output":0.3},"knowledge":"2025-06-25","release_date":"2025-12-15","last_updated":"2025-12-15","open_weights":true},
  "nvidia-nemotron-3-ultra-550b-a55b": {"name":"NVIDIA Nemotron 3 Ultra","description":"Largest Nemotron 3 model for maximum open-weight reasoning and agent accuracy","family":"nemotron","canonical_model_id":"nvidia/nemotron-3-ultra-550b-a55b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":32768},"cost":{"input":0.625,"output":3.125,"cache_read":0.1875},"knowledge":"2025-09","release_date":"2026-06-04","last_updated":"2026-06-04","open_weights":true},
  "nvidia/parakeet-tdt-0.6b-v3": {"name":"nvidia/parakeet-tdt-0.6b-v3","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "olafangensan-glm-4.7-flash-heretic": {"name":"GLM 4.7 Flash Heretic","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":24000},"cost":{"input":0.07,"output":0.4,"cache_read":0.035},"release_date":"2026-02-04","last_updated":"2026-06-11","open_weights":true},
  "openai-gpt-4o-2024-11-20": {"name":"GPT-4o","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt","canonical_model_id":"openai/gpt-4o-2024-11-20","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":2.5,"output":10,"cache_read":1.25},"knowledge":"2023-09","release_date":"2024-11-20","last_updated":"2024-11-20","open_weights":false},
  "openai-gpt-4o-mini-2024-07-18": {"name":"GPT-4o Mini","description":"Small omni GPT for cheap multimodal assistance and production-scale traffic","family":"gpt-mini","canonical_model_id":"openai/gpt-4o-mini","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.6,"cache_read":0.075},"knowledge":"2023-09","release_date":"2024-07-18","last_updated":"2024-07-18","open_weights":false},
  "openai-gpt-52": {"name":"GPT-5.2","description":"Reliable GPT generation for broad coding, writing, and tool-assisted product work","family":"gpt","canonical_model_id":"openai/gpt-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"input":272000,"output":65536},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2025-12-11","last_updated":"2025-12-11","open_weights":false},
  "openai-gpt-53-codex": {"name":"GPT-5.3 Codex","description":"Coding-optimized GPT model for repository edits, reviews, and agentic software work","family":"gpt-codex","canonical_model_id":"openai/gpt-5.3-codex","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":1.75,"output":14,"cache_read":0.175},"knowledge":"2025-08-31","release_date":"2026-02-05","last_updated":"2026-02-05","open_weights":false},
  "openai-gpt-54": {"name":"GPT-5.4","description":"Agent-ready GPT for coding and computer-use workflows at a lower cost","family":"gpt","canonical_model_id":"openai/gpt-5.4","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"input":922000,"output":131072},"cost":{"input":2.5,"output":15,"cache_read":0.25,"tiers":[{"input":5,"output":22.5,"cache_read":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":5,"output":22.5,"cache_read":0.5}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false},
  "openai-gpt-54-mini": {"name":"GPT-5.4 Mini","description":"Strong small GPT for coding subagents, quick tool use, and high-volume work","family":"gpt-mini","canonical_model_id":"openai/gpt-5.4-mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":400000,"input":272000,"output":128000},"cost":{"input":0.75,"output":4.5,"cache_read":0.075},"knowledge":"2025-08-31","release_date":"2026-03-17","last_updated":"2026-03-17","open_weights":false},
  "openai-gpt-54-pro": {"name":"GPT-5.4 Pro","description":"More exact GPT-5.4 tier for demanding professional reasoning and agent tasks","family":"gpt-pro","canonical_model_id":"openai/gpt-5.4-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}},"knowledge":"2025-08-31","release_date":"2026-03-05","last_updated":"2026-03-05","open_weights":false},
  "openai-gpt-55": {"name":"GPT-5.5","description":"Default frontier GPT for coding, computer use, research, and knowledge work","family":"gpt","canonical_model_id":"openai/gpt-5.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"input":922000,"output":131072},"cost":{"input":5,"output":30,"cache_read":0.5,"tiers":[{"input":10,"output":45,"cache_read":1,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":10,"output":45,"cache_read":1}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false},
  "openai-gpt-55-pro": {"name":"GPT-5.5 Pro","description":"Highest-accuracy GPT-5.5 tier for slower, precision-heavy reasoning and coding","family":"gpt-pro","canonical_model_id":"openai/gpt-5.5-pro","reasoning":true,"reasoning_options":[{"type":"effort","values":["medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"input":922000,"output":128000},"cost":{"input":30,"output":180,"tiers":[{"input":60,"output":270,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":60,"output":270}},"knowledge":"2025-12-01","release_date":"2026-04-23","last_updated":"2026-04-23","open_weights":false},
  "openai-gpt-56-luna": {"name":"GPT-5.6 Luna","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "openai-gpt-56-luna-pro": {"name":"GPT-5.6 Luna Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-luna","canonical_model_id":"openai/gpt-5.6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.2,"output":1.2,"cache_read":0.02,"cache_write":0.25,"tiers":[{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.4,"output":1.8,"cache_read":0.04,"cache_write":0.5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "openai-gpt-56-sol": {"name":"GPT-5.6 Sol","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "openai-gpt-56-sol-pro": {"name":"GPT-5.6 Sol Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-sol","canonical_model_id":"openai/gpt-5.6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.4,"cache_write":5,"tiers":[{"input":8,"output":30,"cache_read":0.8,"cache_write":10,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":8,"output":30,"cache_read":0.8,"cache_write":10}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "openai-gpt-56-terra": {"name":"GPT-5.6 Terra","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "openai-gpt-56-terra-pro": {"name":"GPT-5.6 Terra Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-terra","canonical_model_id":"openai/gpt-5.6-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":12,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":18,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-02-16","release_date":"2026-07-09","last_updated":"2026-07-09","open_weights":false},
  "openai-gpt-6-astra": {"name":"GPT-6 Astra","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false},
  "openai-gpt-6-astra-pro": {"name":"GPT-6 Astra Pro","description":"Frontier GPT model for professional reasoning, coding, and multimodal work","family":"gpt-astra","canonical_model_id":"openai/gpt-6-astra","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5,"tiers":[{"input":20,"output":75,"cache_read":2,"cache_write":25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":20,"output":75,"cache_read":2,"cache_write":25}},"knowledge":"2026-04-30","release_date":"2026-09-04","last_updated":"2026-09-04","open_weights":false},
  "openai-gpt-6-luna": {"name":"GPT-6 Luna","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-luna","canonical_model_id":"openai/gpt-6-luna","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":0.2,"output":0.75,"cache_read":0.02,"cache_write":0.25}},"knowledge":"2026-05-18","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "openai-gpt-6-sol": {"name":"GPT-6 Sol","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","canonical_model_id":"openai/gpt-6-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.4,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.4,"cache_write":5}},"knowledge":"2026-04-20","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false},
  "openai-gpt-61-sol": {"name":"GPT-6.1 Sol","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5,"tiers":[{"input":4,"output":15,"cache_read":0.2,"cache_write":5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":4,"output":15,"cache_read":0.2,"cache_write":5}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false},
  "openai-gpt-61-sol-ultrafast": {"name":"GPT-6.1 Sol Ultrafast","description":"GPT model for general reasoning, writing, coding, and tool-assisted tasks","family":"gpt-sol","canonical_model_id":"openai/gpt-6.1-sol","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":12,"output":60,"cache_read":0.6,"cache_write":15,"tiers":[{"input":24,"output":90,"cache_read":1.2,"cache_write":30,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":24,"output":90,"cache_read":1.2,"cache_write":30}},"knowledge":"2026-04-30","release_date":"2026-09-29","last_updated":"2026-09-29","open_weights":false},
  "openai-gpt-61-terra": {"name":"GPT-6.1 Terra","description":"Fast, efficient model for visual reasoning, precise computer use, and clear, natural writing.","family":"gpt-terra","canonical_model_id":"openai/gpt-6.1-terra","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1050000,"input":922000,"output":128000},"cost":{"input":1,"output":5,"cache_read":0.05,"cache_write":1.25,"tiers":[{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5,"tier":{"type":"context","size":272000}}],"context_over_200k":{"input":2,"output":7.5,"cache_read":0.1,"cache_write":2.5}},"knowledge":"2026-06-01","release_date":"2026-10-26","last_updated":"2026-10-26","open_weights":false},
  "openai-gpt-oss-120b": {"name":"OpenAI GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.07,"output":0.3},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true},
  "openai/whisper-large-v3": {"name":"openai/whisper-large-v3","modalities":{"input":["audio"]},"rig":{"pinned":["name","modalities"]}},
  "qwen-3-6-plus": {"name":"Qwen 3.6 Plus Uncensored","description":"Earlier Qwen multimodal workhorse for million-token agent and document tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.6-plus","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.625,"output":3.75,"cache_read":0.0625,"cache_write":0.78,"tiers":[{"input":2.5,"output":7.5,"cache_read":0.0625,"cache_write":0.78,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":2.5,"output":7.5,"cache_read":0.0625,"cache_write":0.78}},"knowledge":"2025-04","release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":false},
  "qwen-3-7-max": {"name":"Qwen 3.7 Max","description":"Qwen frontier model tuned for agent frameworks, coding assistants, and long tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.7-max","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":2.7,"output":8.05,"cache_read":0.27,"cache_write":3.35},"release_date":"2026-05-21","last_updated":"2026-05-21","open_weights":false},
  "qwen-3-7-plus": {"name":"Qwen 3.7 Plus","description":"Multimodal Qwen workhorse for long-context agents, visual inputs, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3.7-plus","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.5,"output":2,"cache_read":0.05,"cache_write":0.625,"tiers":[{"input":1.5,"output":6,"cache_read":0.15,"cache_write":1.875,"tier":{"type":"context","size":256000}}],"context_over_200k":{"input":1.5,"output":6,"cache_read":0.15,"cache_write":1.875}},"knowledge":"2025-04","release_date":"2026-06-02","last_updated":"2026-06-02","open_weights":false},
  "qwen-3-8-2-4t-a95b": {"name":"Qwen 3.8 2.4T","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.8-2.4t-a95b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":2.5,"output":7.5,"cache_read":0.3125},"release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":true},
  "qwen-3-8-27b": {"name":"Qwen 3.8 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":262144,"output":65536},"cost":{"input":0.45,"output":3.2},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "qwen-3-8-flash": {"name":"Qwen 3.8 Flash","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.8-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.14,"output":0.49,"cache_read":0.014},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":false},
  "qwen-3-8-max": {"name":"Qwen 3.8 Max","description":"Preview Qwen flagship for million-token multimodal reasoning and long-horizon agentic workflows","family":"qwen","canonical_model_id":"alibaba/qwen3.8-max-preview","reasoning":true,"reasoning_options":[],"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":2.5,"output":7.5,"cache_read":0.3125,"cache_write":3.125},"release_date":"2026-07-19","last_updated":"2026-07-19","open_weights":false},
  "qwen-image": {"name":"qwen-image","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "qwen3-235b-a22b-instruct-2507": {"name":"Qwen 3 235B A22B Instruct 2507","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-instruct-2507","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.15,"output":0.75},"release_date":"2025-07-21","last_updated":"2025-07-21","open_weights":true},
  "qwen3-235b-a22b-thinking-2507": {"name":"Qwen 3 235B A22B Thinking 2507","description":"Qwen reasoning model for deliberate problem solving, math, and coding","family":"qwen","canonical_model_id":"alibaba/qwen3-235b-a22b-thinking-2507","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.45,"output":3.5},"release_date":"2025-07-25","last_updated":"2025-07-25","open_weights":true},
  "qwen3-5-35b-a3b": {"name":"Qwen 3.5 35B A3B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-35b-a3b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":256000,"output":16384},"cost":{"input":0.15,"output":1,"cache_read":0.05},"release_date":"2026-02-23","last_updated":"2026-02-23","open_weights":true},
  "qwen3-5-397b-a17b": {"name":"Qwen 3.5 397B","description":"Large open Qwen multimodal MoE for visual agents and long technical tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.5-397b-a17b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":0.75,"output":4.5},"release_date":"2026-02-15","last_updated":"2026-02-15","open_weights":true},
  "qwen3-5-9b": {"name":"Qwen 3.5 9B","canonical_model_id":"alibaba/qwen3.5-9b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":256000,"output":32768},"cost":{"input":0.1,"output":0.15},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","limit.output","cost.input","cost.output"]}},
  "qwen3-6-27b": {"name":"Qwen 3.6 27B","description":"Qwen vision-language model for visual reasoning, documents, and agent tasks","family":"qwen","canonical_model_id":"alibaba/qwen3.6-27b","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.325,"output":3.25},"release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true},
  "qwen3-6-35b-a3b": {"name":"Qwen 3.6 35B A3B","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3.6-35b-a3b","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.1,"output":1},"release_date":"2026-04-17","last_updated":"2026-04-17","open_weights":true},
  "qwen3-coder-480b-a35b-instruct-turbo": {"name":"Qwen 3 Coder 480B Turbo","description":"Qwen coding model for software agents, repository edits, and code reasoning","family":"qwen","reasoning":false,"tool_call":true,"structured_output":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.35,"output":1.5,"cache_read":0.04},"release_date":"2026-01-27","last_updated":"2026-06-11","open_weights":true},
  "qwen3-next-80b": {"name":"Qwen 3 Next 80b","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","canonical_model_id":"alibaba/qwen3-next-80b-a3b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":256000,"output":16384},"cost":{"input":0.35,"output":1.9},"knowledge":"2025-04","release_date":"2025-09","last_updated":"2025-09","open_weights":true},
  "qwen3-vl-235b-a22b": {"name":"Qwen3 VL 235B","description":"Multimodal model for analyzing text, images, documents, and rich media","reasoning":false,"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.21,"output":1.9,"cache_read":0.1},"release_date":"2026-01-16","last_updated":"2026-06-11","open_weights":true},
  "seed-2-1-turbo": {"name":"Seed 2.1 Turbo","description":"Multimodal reasoning model for visual analysis, planning, and tool use","family":"seed","canonical_model_id":"bytedance-seed/seed-2.1-turbo","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":256000,"output":65536},"cost":{"input":0.625,"output":3.125,"cache_read":0.125},"release_date":"2026-06-23","last_updated":"2026-06-23","open_weights":false},
  "text-embedding-bge-en-icl": {"name":"text-embedding-bge-en-icl","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "text-embedding-bge-m3": {"name":"text-embedding-bge-m3","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "text-embedding-multilingual-e5-large-instruct": {"name":"text-embedding-multilingual-e5-large-instruct","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "text-embedding-qwen3-0-6b": {"name":"text-embedding-qwen3-0-6b","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "text-embedding-qwen3-8b": {"name":"text-embedding-qwen3-8b","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "tts-elevenlabs-turbo-v2-5": {"name":"tts-elevenlabs-turbo-v2-5","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "tts-inworld-1-5-max": {"name":"tts-inworld-1-5-max","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "tts-kokoro": {"name":"tts-kokoro","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "tts-xai-v1": {"name":"tts-xai-v1","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "venice-sd35": {"name":"venice-sd35","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "venice-uncensored-1-2": {"name":"Venice Uncensored 1.2","description":"Multimodal model for analyzing text, images, documents, and rich media","family":"venice","reasoning":false,"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.2,"output":0.9},"release_date":"2026-04-01","last_updated":"2026-06-11","open_weights":true},
  "venice-uncensored-role-play": {"name":"Venice Role Play Uncensored","description":"Multimodal model for analyzing text, images, documents, and rich media","family":"venice","reasoning":false,"tool_call":true,"structured_output":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":4096},"cost":{"input":0.5,"output":2},"release_date":"2026-02-20","last_updated":"2026-06-11","open_weights":true},
  "xiaomi-mimo-v2-5": {"name":"MiMo-V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":65536},"cost":{"input":0.4,"output":2,"cache_read":0.08},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true},
  "xiaomi-mimo-v2-6-flash": {"name":"MiMo-V2.6-Flash","description":"MiMo flash model for fast multimodal assistance and agent workflows","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.175,"output":0.35,"cache_read":0.00375},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true},
  "z-ai-glm-5-3": {"name":"GLM 5.3","description":"Flagship GLM model for hybrid reasoning, coding, and agentic engineering","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.75,"output":5.5,"cache_read":0.325},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "z-ai-glm-5-3-flash": {"name":"GLM 5.3 Flash","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true},
  "z-ai-glm-5-turbo": {"name":"GLM 5 Turbo","description":"Faster GLM-5 lane for coding agents that need lower latency","family":"glm","canonical_model_id":"zhipuai/glm-5-turbo","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":32768},"cost":{"input":1.2,"output":4,"cache_read":0.24},"release_date":"2026-03-16","last_updated":"2026-03-16","open_weights":false},
  "z-ai-glm-5v-turbo": {"name":"GLM 5V Turbo","description":"Fast GLM vision model for screenshots, documents, and multimodal agent tasks","family":"glm","canonical_model_id":"zhipuai/glm-5v-turbo","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":32768},"cost":{"input":1.5,"output":5,"cache_read":0.3},"release_date":"2026-04-01","last_updated":"2026-04-01","open_weights":false},
  "z-image-turbo": {"name":"z-image-turbo","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "zai-org-glm-4.6": {"name":"GLM 4.6","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":198000,"output":16384},"cost":{"input":0.43,"output":1.75,"cache_read":0.08},"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","open_weights":true},
  "zai-org-glm-4.7": {"name":"GLM 4.7","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":198000,"output":16384},"cost":{"input":0.55,"output":2.65,"cache_read":0.11},"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","open_weights":true},
  "zai-org-glm-5": {"name":"GLM 5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":198000,"output":32000},"cost":{"input":1,"output":3.2,"cache_read":0.2},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true},
  "zai-org-glm-5-1": {"name":"GLM 5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","canonical_model_id":"zhipuai/glm-5.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":80000},"cost":{"input":1.54,"output":4.84,"cache_read":0.286},"release_date":"2026-04-07","last_updated":"2026-04-07","open_weights":true},
  "zai-org-glm-5-2": {"name":"GLM 5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true}
}},
"vertexai": {"models": {
  "claude-fable-5-1@default": {"name":"Claude Fable 5.1","description":"Claude model for demanding reasoning and long-horizon agentic work","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5-1","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":0.25,"cache_write":12.5},"knowledge":"2026-06","release_date":"2026-09-01","last_updated":"2026-09-01","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never"}},
  "claude-fable-5@default": {"name":"Claude Fable 5","description":"Claude model for creative writing, analysis, and controlled agent workflows","family":"claude-fable","canonical_model_id":"anthropic/claude-fable-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":10,"output":50,"cache_read":1,"cache_write":12.5},"knowledge":"2026-01-31","release_date":"2026-06-09","last_updated":"2026-06-09","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-haiku-4-5@20251001": {"name":"Claude Haiku 4.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-4-5-20251001","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":1,"output":5,"cache_read":0.1,"cache_write":1.25},"knowledge":"2025-02-28","release_date":"2025-10-15","last_updated":"2025-10-15","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-haiku-5-5@default": {"name":"Claude Haiku 5.5","description":"Fast Claude model for responsive assistance, classification, and lightweight agents","family":"claude-haiku","canonical_model_id":"anthropic/claude-haiku-5-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":0.1,"output":0.5,"cache_read":0.01,"cache_write":0.125,"tiers":[{"input":0.5,"output":2.5,"cache_read":0.05,"cache_write":0.625,"tier":{"type":"context","size":100000}}]},"knowledge":"2026-06","release_date":"2026-10-07","last_updated":"2026-10-07","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"pinned":["reasoning_options"]}},
  "claude-opus-4-1@20250805": {"name":"Claude Opus 4.1","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-1-20250805","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"knowledge":"2025-03-31","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-opus-4-5@20251101": {"name":"Claude Opus 4.5","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-5-20251101","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05","release_date":"2025-11-24","last_updated":"2025-11-24","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-opus-4-6@default": {"name":"Claude Opus 4.6","description":"High-end Claude for difficult coding, planning, and slower expert reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2025-05-31","release_date":"2026-02-05","last_updated":"2026-03-13","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "claude-opus-4-7@default": {"name":"Claude Opus 4.7","description":"Stronger Opus tier for advanced software work and high-stakes reasoning","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01-31","release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-opus-4-8@default": {"name":"Claude Opus 4.8","description":"Top Claude Opus tier for the hardest reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-8","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-opus-4@20250514": {"name":"Claude Opus 4","description":"Flagship Claude model for deep reasoning, coding, and long-horizon agents","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-4-20250514","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":32000},"cost":{"input":15,"output":75,"cache_read":1.5,"cache_write":18.75},"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-opus-5-5@default": {"name":"Claude Opus 5.5","description":"Claude model for long-running agentic coding and knowledge work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":4,"output":20,"cache_read":0.2,"cache_write":5},"knowledge":"2026-06","release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"medium","rejects_forced_tool_choice":true,"sampling":"never"}},
  "claude-opus-5@default": {"name":"Claude Opus 5","description":"Strongest Claude Opus model for coding, agents, and professional work","family":"claude-opus","canonical_model_id":"anthropic/claude-opus-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":5,"output":25,"cache_read":0.5,"cache_write":6.25},"knowledge":"2026-05","release_date":"2026-07-24","last_updated":"2026-07-24","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"sampling":"never"}},
  "claude-sonnet-4-5@20250929": {"name":"Claude Sonnet 4.5","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-5-20250929","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-07-31","release_date":"2025-09-29","last_updated":"2025-09-29","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-sonnet-4-6@default": {"name":"Claude Sonnet 4.6","description":"Claude workhorse for coding agents, careful analysis, and production cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","max"]},{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-08-31","release_date":"2026-02-17","last_updated":"2026-03-13","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"]}},
  "claude-sonnet-4@20250514": {"name":"Claude Sonnet 4","description":"Balanced Claude model for coding, analysis, agent workflows, and cost control","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-4-20250514","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":1024},{"type":"toggle"}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":200000,"output":64000},"cost":{"input":3,"output":15,"cache_read":0.3,"cache_write":3.75},"knowledge":"2025-03-31","release_date":"2025-05-22","last_updated":"2025-05-22","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"cache":["none","short","long"],"pinned":["reasoning_options"],"prefill":true}},
  "claude-sonnet-5-5@default": {"name":"Claude Sonnet 5.5","description":"Fast Claude model for everyday coding, agents, and knowledge work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5-5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh","max"]},{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.1,"cache_write":2.5},"knowledge":"2026-06","release_date":"2026-09-28","last_updated":"2026-09-28","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"binds_context":true,"cache":["none","short","long"],"mid_conversation_system":true,"pinned":["reasoning_options"],"reasoning_default":"high","rejects_forced_tool_choice":true,"sampling":"never","thinking_off":"between_tools"}},
  "claude-sonnet-5@default": {"name":"Claude Sonnet 5","description":"Everyday Claude agent model for coding, planning, browsing, and general work","family":"claude-sonnet","canonical_model_id":"anthropic/claude-sonnet-5","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["low","medium","high","xhigh","max"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":128000},"cost":{"input":2,"output":10,"cache_read":0.2,"cache_write":2.5},"knowledge":"2026-01-31","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false,"provider":{"npm":"@ai-sdk/google-vertex/anthropic"},"rig":{"adaptive_thinking":true,"cache":["none","short","long"],"pinned":["reasoning_options"],"sampling":"never"}},
  "codestral-2": {"name":"Codestral 2","description":"Mistral code model for low-latency completion, fill-in-the-middle, and code generation","family":"codestral","canonical_model_id":"mistral/codestral-2508","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.3,"output":0.9},"release_date":"2025-07-30","last_updated":"2025-07-30","open_weights":false,"provider":{"npm":"@ai-sdk/mistral","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/publishers/mistralai/models"}},
  "deepseek-ai/deepseek-v3.1-maas": {"name":"DeepSeek V3.1","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":32768},"cost":{"input":0.6,"output":1.7,"cache_read":0.06},"release_date":"2025-08-28","last_updated":"2025-08-28","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "deepseek-ai/deepseek-v3.2-maas": {"name":"DeepSeek V3.2","description":"DeepSeek chat model for instruction following, coding, and analysis","family":"deepseek","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":163840,"output":65536},"cost":{"input":0.56,"output":1.68,"cache_read":0.056},"release_date":"2025-12-17","last_updated":"2026-04-04","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "gemini-2.5-flash": {"name":"Gemini 2.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":0,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03,"input_audio":1},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-flash-image": {"name":"Nano Banana","description":"Nano Banana image model for fast generation, edits, and character-consistent assets","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-2.5-flash-image","reasoning":false,"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text","image"]},"limit":{"context":32768,"output":32768},"cost":{"input":0.3,"output":30},"knowledge":"2024-06","release_date":"2025-08-26","last_updated":"2025-08-26","open_weights":false},
  "gemini-2.5-flash-lite": {"name":"Gemini 2.5 Flash-Lite","description":"Lean Gemini 2.5 lane for cheap multimodal traffic and quick agents","family":"gemini-flash-lite","canonical_model_id":"google/gemini-2.5-flash-lite","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"budget_tokens","min":512,"max":24576}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.1,"output":0.4,"cache_read":0.01,"input_audio":0.3},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-flash-tts": {"name":"Gemini 2.5 Flash TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gemini-flash","type":"speech","canonical_model_id":"google/gemini-2.5-flash-tts","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":32768,"input":8192,"output":16384},"cost":{"input":0.5,"output":10},"knowledge":"2025-01","release_date":"2025-09-30","last_updated":"2025-12-10","open_weights":false},
  "gemini-2.5-pro": {"name":"Gemini 2.5 Pro","description":"Google's proven reasoning model for coding, math, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-2.5-pro","reasoning":true,"reasoning_options":[{"type":"budget_tokens","min":128,"max":32768}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","audio","video","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.25,"output":10,"cache_read":0.125,"tiers":[{"input":2.5,"output":15,"cache_read":0.25,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":15,"cache_read":0.25}},"knowledge":"2025-01","release_date":"2025-06-17","last_updated":"2025-06-17","open_weights":false,"status":"deprecated"},
  "gemini-2.5-pro-tts": {"name":"Gemini 2.5 Pro TTS","description":"Speech generation model for controllable voice, narration, and audio delivery","family":"gemini-pro","type":"speech","canonical_model_id":"google/gemini-2.5-pro-tts","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["audio"]},"limit":{"context":32768,"input":8192,"output":16384},"cost":{"input":1,"output":20},"knowledge":"2025-01","release_date":"2025-09-30","last_updated":"2025-12-10","open_weights":false},
  "gemini-3-flash-preview": {"name":"Gemini 3 Flash Preview","description":"New Gemini flash lane bringing frontier-style multimodal reasoning to cheaper runs","family":"gemini-flash","canonical_model_id":"google/gemini-3-flash-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.5,"output":3,"cache_read":0.05,"input_audio":1},"knowledge":"2025-01","release_date":"2025-12-17","last_updated":"2025-12-17","open_weights":false,"status":"beta"},
  "gemini-3-pro-image": {"name":"Nano Banana Pro","description":"Nano Banana Pro for higher-fidelity image generation and design-heavy edits","family":"gemini-pro","type":"image","canonical_model_id":"google/gemini-3-pro-image","reasoning":true,"reasoning_options":[],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text","image"]},"limit":{"context":65536,"output":32768},"cost":{"input":2,"output":120,"cache_read":0.2},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "gemini-3.1-flash-image": {"name":"Nano Banana 2","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-3.1-flash-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"tool_call":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.5,"output":60,"cache_read":0.05},"knowledge":"2025-01","release_date":"2026-05-28","last_updated":"2026-05-28","open_weights":false},
  "gemini-3.1-flash-lite": {"name":"Gemini 3.1 Flash Lite","description":"Low-latency Gemini model for high-volume multimodal and agent workloads","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.1-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.25,"output":1.5,"cache_read":0.025,"input_audio":0.5},"knowledge":"2025-01","release_date":"2026-05-07","last_updated":"2026-05-07","open_weights":false},
  "gemini-3.1-flash-lite-image": {"name":"Nano Banana 2 Lite","description":"Fastest, most cost-efficient Gemini image model for high-volume 1K generation and editing","family":"gemini-flash-lite","type":"image","canonical_model_id":"google/gemini-3.1-flash-lite-image","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","high"]}],"tool_call":true,"structured_output":false,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text","image"]},"limit":{"context":65536,"output":4096},"cost":{"input":0.25,"output":30,"cache_read":0.025},"knowledge":"2025-01","release_date":"2026-06-30","last_updated":"2026-06-30","open_weights":false},
  "gemini-3.1-pro-preview": {"name":"Gemini 3.1 Pro Preview","description":"Reasoning-first Gemini preview for agentic coding and complex problem solving","family":"gemini-pro","canonical_model_id":"google/gemini-3.1-pro-preview","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false,"status":"beta"},
  "gemini-3.1-pro-preview-customtools": {"name":"Gemini 3.1 Pro Preview Custom Tools","description":"Advanced Gemini model for complex reasoning, coding, and multimodal analysis","family":"gemini-pro","canonical_model_id":"google/gemini-3.1-pro-preview-customtools","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":12,"cache_read":0.2,"tiers":[{"input":4,"output":18,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":18,"cache_read":0.4}},"knowledge":"2025-01","release_date":"2026-02-19","last_updated":"2026-02-19","open_weights":false,"status":"beta"},
  "gemini-3.5-flash": {"name":"Gemini 3.5 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.5-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":1.5,"output":9,"cache_read":0.15,"input_audio":1.5},"knowledge":"2025-01","release_date":"2026-05-19","last_updated":"2026-05-19","open_weights":false},
  "gemini-3.5-flash-lite": {"name":"Gemini 3.5 Flash Lite","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash-lite","canonical_model_id":"google/gemini-3.5-flash-lite","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.3,"output":2.5,"cache_read":0.03},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false},
  "gemini-3.6-flash": {"name":"Gemini 3.6 Flash","description":"Fast Gemini model balancing multimodal reasoning, tool use, and cost","family":"gemini-flash","canonical_model_id":"google/gemini-3.6-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-07-21","last_updated":"2026-07-21","open_weights":false,"status":"deprecated"},
  "gemini-3.7-flash": {"name":"Gemini 3.7 Flash","description":"High-efficiency Gemini model for agentic workflows, coding, and multimodal reasoning","family":"gemini-flash","canonical_model_id":"google/gemini-3.7-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"knowledge":"2026-03","release_date":"2026-08-13","last_updated":"2026-08-13","open_weights":false},
  "gemini-3.8-flash": {"name":"Gemini 3.8 Flash","description":"Google's most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows","family":"gemini-flash","canonical_model_id":"google/gemini-3.8-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":0.75,"output":3.75,"cache_read":0.075,"input_audio":0.75},"release_date":"2026-09-02","last_updated":"2026-09-02","open_weights":false},
  "gemini-4-argon": {"name":"Gemini 4 Argon","description":"Google's flagship reasoning model for complex software engineering, deep math and science, and long-horizon autonomous agents","family":"gemini","canonical_model_id":"google/gemini-4-argon","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":false,"attachment":true,"modalities":{"input":["text","image","video","audio","pdf"],"output":["text"]},"limit":{"context":1048576,"output":65536},"cost":{"input":2,"output":10,"cache_read":0.1,"input_audio":2},"knowledge":"2026-03","release_date":"2026-10-13","last_updated":"2026-10-13","open_weights":false},
  "gemini-embedding-001": {"name":"Gemini Embedding 001","description":"Embedding model for semantic search, retrieval, clustering, and ranking pipelines","family":"gemini","type":"embedding","canonical_model_id":"google/gemini-embedding-001","reasoning":false,"tool_call":false,"temperature":false,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":2048,"output":1},"cost":{"input":0.15,"output":0},"knowledge":"2025-05","release_date":"2025-05-20","last_updated":"2025-05-20","open_weights":false},
  "gemini-nano-banana-2.1": {"name":"Nano Banana 2.1","description":"Image model for prompt-driven generation, multi-turn editing, and visual design workflows","family":"gemini-flash","type":"image","canonical_model_id":"google/gemini-nano-banana-2.1","reasoning":true,"reasoning_options":[{"type":"effort","values":["minimal","medium","high"]}],"tool_call":false,"structured_output":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","video","pdf"],"output":["text","image"]},"limit":{"context":131072,"output":32768},"cost":{"input":1.5,"output":30,"cache_read":0.15},"release_date":"2026-10-06","last_updated":"2026-10-06","open_weights":false},
  "google/gemma-4-26b-a4b-it-maas": {"name":"Gemma 4 26B A4B IT","description":"Open Gemma instruction model for efficient chat and self-hosted deployments","family":"gemma","canonical_model_id":"google/gemma-4-26b-a4b-it","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":262144,"output":128000},"cost":{"input":0.15,"output":0.6,"cache_read":0.015},"release_date":"2026-04-02","last_updated":"2026-04-02","open_weights":true,"status":"alpha","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "meta/llama-3.3-70b-instruct-maas": {"name":"Llama 3.3 70B Instruct","description":"Open Llama instruction model for multilingual chat, reasoning, and coding","family":"llama","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":128000,"output":8192},"cost":{"input":0.72,"output":0.72},"knowledge":"2023-12","release_date":"2025-04-29","last_updated":"2025-04-29","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "meta/llama-4-maverick-17b-128e-instruct-maas": {"name":"Llama 4 Maverick 17B 128E Instruct","description":"Open multimodal Llama model for strong reasoning and fast responses","family":"llama","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":524288,"output":8192},"cost":{"input":0.35,"output":1.15},"knowledge":"2024-08","release_date":"2025-04-29","last_updated":"2025-04-29","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "meta/llama-4-scout-17b-16e-instruct-maas": {"name":"Llama 4 Scout 17B Instruct","description":"Open Llama with long-context vision for efficient multimodal agents","family":"llama","canonical_model_id":"meta/llama-4-scout-17b-instruct","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":1310720,"output":8192},"cost":{"input":0.25,"output":0.7},"knowledge":"2024-08","release_date":"2025-04-05","last_updated":"2025-04-05","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "mistral-medium-3": {"name":"Mistral Medium 3","description":"Mistral model for multilingual chat, reasoning, and tool-assisted workflows","family":"mistral-medium","canonical_model_id":"mistral/mistral-medium-2505","reasoning":false,"tool_call":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.4,"output":2},"knowledge":"2025-05","release_date":"2025-05-07","last_updated":"2025-05-07","open_weights":false,"provider":{"npm":"@ai-sdk/mistral","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/publishers/mistralai/models"}},
  "mistral-small-2503": {"name":"Mistral Small 3.1 24B","description":"Efficient multimodal model for instruction following, coding, reasoning, and function calling","family":"mistral-small","canonical_model_id":"mistral/mistral-small-3-1-24b-instruct-2503","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":16384},"cost":{"input":0.1,"output":0.3},"knowledge":"2024-06","release_date":"2025-03-17","last_updated":"2025-03-17","open_weights":true,"provider":{"npm":"@ai-sdk/mistral","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/publishers/mistralai/models"}},
  "moonshotai/kimi-k2-thinking-maas": {"name":"Kimi K2 Thinking","description":"Kimi reasoning model for long-horizon research, planning, and tool use","family":"kimi-thinking","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":262144},"cost":{"input":0.6,"output":2.5,"cache_read":0.06},"knowledge":"2024-08","release_date":"2025-11-13","last_updated":"2025-11-13","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "openai/gpt-oss-120b-maas": {"name":"GPT OSS 120B","description":"Open GPT reasoning model for self-hosted agents and controllable deployments","family":"gpt-oss","canonical_model_id":"openai/gpt-oss-120b","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":131072},"cost":{"input":0.09,"output":0.36},"knowledge":"2024-06","release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "openai/gpt-oss-20b-maas": {"name":"GPT OSS 20B","description":"Open-weight GPT model for self-hosted reasoning and instruction-following workloads","family":"gpt-oss","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":32768},"cost":{"input":0.07,"output":0.25,"cache_read":0.007},"release_date":"2025-08-05","last_updated":"2025-08-05","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "qwen/qwen3-235b-a22b-instruct-2507-maas": {"name":"Qwen3 235B A22B Instruct","description":"Qwen instruction model for multilingual chat, reasoning, and tool use","family":"qwen","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"modalities":{"input":["text"],"output":["text"]},"limit":{"context":262144,"output":16384},"cost":{"input":0.22,"output":0.88},"release_date":"2025-08-13","last_updated":"2025-08-13","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "xai/grok-4.1-fast-non-reasoning": {"name":"Grok 4.1 Fast","description":"Fast Grok model for responsive chat, tool-assisted work, and low-latency responses","family":"grok","canonical_model_id":"xai/grok-4.1-fast","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":30000},"cost":{"input":0.2,"output":0.5,"cache_read":0.05},"release_date":"2025-11-19","last_updated":"2025-11-19","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "xai/grok-4.1-fast-reasoning": {"name":"Grok 4.1 Fast (Reasoning)","description":"Fast Grok model for responsive chat, reasoning, and tool-assisted work","family":"grok","canonical_model_id":"xai/grok-4.1-fast-reasoning","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":128000,"output":30000},"cost":{"input":0.2,"output":0.5,"cache_read":0.05},"release_date":"2025-11-19","last_updated":"2025-11-19","open_weights":false,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "xai/grok-4.20-non-reasoning": {"name":"Grok 4.20 (Non-Reasoning)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","canonical_model_id":"xai/grok-4.20-0309-non-reasoning","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":2000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-03-09","last_updated":"2026-03-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "xai/grok-4.20-reasoning": {"name":"Grok 4.20 (Reasoning)","description":"Reasoning Grok for document-heavy analysis and long-horizon tool use","family":"grok","canonical_model_id":"xai/grok-4.20-0309-reasoning","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":2000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-03-09","last_updated":"2026-03-09","open_weights":false,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "xai/grok-4.3": {"name":"Grok 4.3","description":"xAI's default Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.3","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":200000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-04-17","last_updated":"2026-04-17","open_weights":false,"status":"beta","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "xai/grok-4.6": {"name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":524288,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":false,"provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "xai/grok-4.7": {"name":"Grok 4.7","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image"],"output":["text"]},"limit":{"context":524288,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false,"status":"beta","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "zai-org/glm-4.7-maas": {"name":"GLM-4.7","description":"GLM model for coding, tool use, and complex reasoning","family":"glm","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":128000},"cost":{"input":0.6,"output":2.2,"cache_read":0.06},"knowledge":"2025-04","release_date":"2026-01-06","last_updated":"2026-01-06","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "zai-org/glm-5-maas": {"name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":128000},"cost":{"input":1,"output":3.2,"cache_read":0.1},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true,"status":"deprecated","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}},
  "zai-org/glm-5.2-maas": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":64000},"cost":{"input":1.4,"output":4.4,"cache_read":0.14},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true,"status":"beta","provider":{"npm":"@ai-sdk/openai-compatible","api":"https://${GOOGLE_VERTEX_ENDPOINT}/v1/projects/${GOOGLE_VERTEX_PROJECT}/locations/${GOOGLE_VERTEX_LOCATION}/endpoints/openapi"}}
}},
"voyageai": {"models": {
  "rerank-1": {"name":"rerank-1","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "rerank-2": {"name":"rerank-2","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "rerank-2-lite": {"name":"rerank-2-lite","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "rerank-2.5": {"name":"rerank-2.5","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "rerank-2.5-lite": {"name":"rerank-2.5-lite","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "rerank-lite-1": {"name":"rerank-lite-1","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "voyage-3-large": {"name":"voyage-3-large","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "voyage-3.5": {"name":"voyage-3.5","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "voyage-code-2": {"name":"voyage-code-2","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "voyage-code-3": {"name":"voyage-code-3","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "voyage-finance-2": {"name":"voyage-finance-2","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "voyage-law-2": {"name":"voyage-law-2","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "voyage.3-5.lite": {"name":"voyage.3-5.lite","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}}
}},
"xai": {"models": {
  "grok-2-1212": {"name":"Grok 2","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":131072},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","status"]}},
  "grok-2-image-1212": {"name":"Grok 2 Image","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}},
  "grok-2-vision-1212": {"name":"Grok 2 Vision","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":32768},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","status"]}},
  "grok-3": {"name":"Grok 3","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":131072},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","status"]}},
  "grok-3-fast": {"name":"Grok 3 Fast","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":131072},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","tool_call","structured_output","temperature","modalities","limit.context","status"]}},
  "grok-3-mini": {"name":"Grok 3 Mini","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":131072},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","status"]}},
  "grok-3-mini-fast": {"name":"Grok 3 Mini Fast","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high"]}],"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text"]},"limit":{"context":131072},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","status"]}},
  "grok-4-0709": {"name":"Grok 4","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"modalities":{"input":["text","image"]},"limit":{"context":256000},"status":"deprecated","rig":{"cache":["short"],"pinned":["name","reasoning","reasoning_options","tool_call","structured_output","temperature","modalities","limit.context","status"],"reasoning_control":"none"}},
  "grok-4.20-0309-non-reasoning": {"name":"Grok 4.20 (Non-Reasoning)","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":false,"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-03-09","last_updated":"2026-03-09","open_weights":false,"rig":{"cache":["short"]}},
  "grok-4.20-0309-reasoning": {"name":"Grok 4.20 (Reasoning)","description":"Reasoning Grok for document-heavy analysis and long-horizon tool use","family":"grok","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-03-09","last_updated":"2026-03-09","open_weights":false,"rig":{"cache":["short"]}},
  "grok-4.20-multi-agent-0309": {"name":"Grok 4.20 Multi-Agent","description":"Grok model for agentic tool use, reasoning, coding, and live assistance","family":"grok","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":false,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-03-09","last_updated":"2026-03-09","open_weights":false,"rig":{"cache":["short"]}},
  "grok-4.3": {"name":"Grok 4.3","description":"xAI's Grok for chat, coding, agentic tools, and lower hallucination risk","family":"grok","reasoning":true,"reasoning_options":[{"type":"effort","values":["none","low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":1000000,"output":30000},"cost":{"input":1.25,"output":2.5,"cache_read":0.2,"tiers":[{"input":2.5,"output":5,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2.5,"output":5,"cache_read":0.4}},"release_date":"2026-04-17","last_updated":"2026-04-17","open_weights":false,"rig":{"cache":["short"],"pinned":["reasoning_options"],"reasoning_default":"low"}},
  "grok-4.5": {"name":"Grok 4.5","description":"xAI's Grok model for chat, coding, agentic tools, and lower hallucination risk","family":"grok","canonical_model_id":"xai/grok-4.5","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.3,"tiers":[{"input":4,"output":12,"cache_read":0.6,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":0.6}},"release_date":"2026-07-08","last_updated":"2026-07-08","open_weights":false,"rig":{"cache":["short"]}},
  "grok-4.6": {"name":"Grok 4.6","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.6","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-02-01","release_date":"2026-08-12","last_updated":"2026-08-12","open_weights":false,"rig":{"cache":["short"]}},
  "grok-4.7": {"name":"Grok 4.7","description":"xAI's frontier model for long-running agents, coding, knowledge work, and visual projects","family":"grok","canonical_model_id":"xai/grok-4.7","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","medium","high","xhigh"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":500000,"output":500000},"cost":{"input":2,"output":6,"cache_read":0.5,"tiers":[{"input":4,"output":12,"cache_read":1,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":4,"output":12,"cache_read":1}},"knowledge":"2026-05","release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":false,"rig":{"cache":["short"]}},
  "grok-build-0.1": {"name":"Grok Build 0.1","description":"Fast Grok coding model tuned for agentic engineering and iterative edits","family":"grok-build","reasoning":true,"reasoning_options":[],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"modalities":{"input":["text","image","pdf"],"output":["text"]},"limit":{"context":256000,"output":256000},"cost":{"input":1,"output":2,"cache_read":0.2,"tiers":[{"input":2,"output":4,"cache_read":0.4,"tier":{"type":"context","size":200000}}],"context_over_200k":{"input":2,"output":4,"cache_read":0.4}},"release_date":"2026-04-16","last_updated":"2026-04-16","open_weights":false,"rig":{"cache":["short"]}},
  "grok-imagine-image": {"name":"Grok Imagine Image","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"grok","type":"image","canonical_model_id":"xai/grok-imagine-image","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["image"]},"limit":{"context":16000,"output":0},"release_date":"2026-01-28","last_updated":"2026-01-28","open_weights":false},
  "grok-imagine-image-pro": {"name":"Grok Imagine Image Pro","modalities":{"input":["text","image"]},"rig":{"pinned":["name","modalities"]}},
  "grok-imagine-image-quality": {"name":"Grok Imagine Image Quality","description":"Higher-fidelity Grok Imagine image model for prompt-driven generation, editing, and visual design workflows","family":"grok","type":"image","canonical_model_id":"xai/grok-imagine-image-quality","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["image"]},"limit":{"context":16000,"output":0},"release_date":"2026-04-03","last_updated":"2026-04-03","open_weights":false,"status":"deprecated"},
  "grok-imagine-video": {"name":"Grok Imagine Video","description":"Image model for prompt-driven generation, editing, and visual design workflows","family":"grok","type":"video","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","video"],"output":["video"]},"limit":{"context":1024,"output":0},"release_date":"2026-01-28","last_updated":"2026-01-28","open_weights":false},
  "grok-imagine-video-1.5": {"name":"Grok Imagine Video 1.5","description":"Video model for text-to-video, image-to-video, and reference-to-video generation with optional voice references","family":"grok","type":"video","canonical_model_id":"xai/grok-imagine-video-1.5","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image","audio"],"output":["video"]},"limit":{"context":1024,"output":0},"release_date":"2026-05-30","last_updated":"2026-05-30","open_weights":false},
  "grok-imagine-video-1.5-lite": {"name":"Grok Imagine Video 1.5 Lite","description":"Cost-efficient video model for text-to-video and image-to-video generation","family":"grok","type":"video","canonical_model_id":"xai/grok-imagine-video-1.5-lite","reasoning":false,"tool_call":false,"temperature":false,"attachment":true,"modalities":{"input":["text","image"],"output":["video"]},"limit":{"context":1024,"output":0},"release_date":"2026-10-01","last_updated":"2026-10-01","open_weights":false},
  "tts-1": {"name":"xAI TTS","modalities":{"input":["text"]},"rig":{"pinned":["name","modalities"]}}
}},
"xiaomimimo": {"models": {
  "mimo-v2.5": {"name":"MiMo-V2.5","description":"Open MiMo model for multimodal coding agents and long-context automation","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"status":"deprecated"},
  "mimo-v2.5-pro": {"name":"MiMo-V2.5-Pro","description":"Stronger MiMo Pro tier for multimodal reasoning and coding-agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5-pro","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.435,"output":0.87,"cache_read":0.0036},"knowledge":"2024-12","release_date":"2026-04-22","last_updated":"2026-04-22","open_weights":true,"status":"deprecated"},
  "mimo-v2.5-pro-ultraspeed": {"name":"MiMo-V2.5-Pro-UltraSpeed","description":"MiMo pro model for strong multimodal reasoning and agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.5-pro-ultraspeed","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":1.305,"output":2.61,"cache_read":0.0108},"knowledge":"2024-12","release_date":"2026-06-08","last_updated":"2026-06-09","open_weights":true,"status":"deprecated"},
  "mimo-v2.6-flash": {"name":"MiMo-V2.6-Flash","description":"MiMo Flash model for multimodal coding agents and long-context automation","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-flash","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.14,"output":0.28,"cache_read":0.0028},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true},
  "mimo-v2.6-pro": {"name":"MiMo-V2.6-Pro","description":"MiMo Pro model for multimodal coding agents and long-context automation","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-pro","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":0.435,"output":0.87,"cache_read":0.0036},"release_date":"2026-09-22","last_updated":"2026-09-22","open_weights":true},
  "mimo-v2.6-pro-ultraspeed": {"name":"MiMo-V2.6-Pro-UltraSpeed","description":"MiMo pro model for strong multimodal reasoning and agent execution","family":"mimo","canonical_model_id":"xiaomi/mimo-v2.6-pro-ultraspeed","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","audio","video"],"output":["text"]},"limit":{"context":1048576,"output":131072},"cost":{"input":4.35,"output":8.7,"cache_read":0.036},"release_date":"2026-09-21","last_updated":"2026-09-21","open_weights":true}
}},
"zai": {"models": {
  "glm-4.5": {"name":"GLM-4.5","description":"Hybrid-reasoning GLM release that made the 4.5 line broadly useful","family":"glm","canonical_model_id":"zhipuai/glm-4.5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":98304},"cost":{"input":0.6,"output":2.2,"cache_read":0.11,"cache_write":0},"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":true},
  "glm-4.5-air": {"name":"GLM-4.5-Air","description":"Lighter GLM-4.5 variant for fast coding assistance and cheaper agents","family":"glm-air","canonical_model_id":"zhipuai/glm-4.5-air","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":98304},"cost":{"input":0.2,"output":1.1,"cache_read":0.03,"cache_write":0},"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":true},
  "glm-4.5-airx": {"name":"GLM-4.5-AIRX","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","modalities"]}},
  "glm-4.5-flash": {"name":"GLM-4.5-Flash","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.5-flash","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":131072,"output":98304},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0},"knowledge":"2025-04","release_date":"2025-07-28","last_updated":"2025-07-28","open_weights":false},
  "glm-4.5v": {"name":"GLM-4.5V","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","canonical_model_id":"zhipuai/glm-4.5v","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":64000,"output":16384},"cost":{"input":0.6,"output":1.8,"cache_read":0.11,"cache_write":0},"knowledge":"2025-04","release_date":"2025-08-11","last_updated":"2025-08-11","open_weights":true},
  "glm-4.6": {"name":"GLM-4.6","description":"Late GLM-4 workhorse for coding agents, reasoning, and structured tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.6","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.11,"cache_write":0},"knowledge":"2025-04","release_date":"2025-09-30","last_updated":"2025-09-30","open_weights":true},
  "glm-4.6-air": {"name":"GLM-4.6-AIR","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","modalities"]}},
  "glm-4.6-x": {"name":"GLM-4.6-X","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"modalities":{"input":["text"]},"rig":{"pinned":["name","reasoning","reasoning_options","tool_call","modalities"]}},
  "glm-4.6v": {"name":"GLM-4.6V","description":"GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm","canonical_model_id":"zhipuai/glm-4.6v","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":0.3,"output":0.9,"cache_read":0.05,"cache_write":0},"knowledge":"2025-04","release_date":"2025-12-08","last_updated":"2025-12-08","open_weights":true},
  "glm-4.6v-flash": {"name":"GLM-4.6V-Flash","description":"Lightweight GLM vision model for visual reasoning, documents, and multimodal agents","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.6v-flash","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video"],"output":["text"]},"limit":{"context":128000,"output":32768},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0},"release_date":"2025-12-08","last_updated":"2025-12-08","open_weights":true},
  "glm-4.7": {"name":"GLM-4.7","description":"Mature GLM model for dependable coding, reasoning, and structured agent tasks","family":"glm","canonical_model_id":"zhipuai/glm-4.7","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":0.6,"output":2.2,"cache_read":0.11,"cache_write":0},"knowledge":"2025-04","release_date":"2025-12-22","last_updated":"2025-12-22","open_weights":true},
  "glm-4.7-flash": {"name":"GLM-4.7-Flash","description":"Budget GLM lane for fast coding help, routing, and everyday automation","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.7-flash","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":131072},"cost":{"input":0,"output":0,"cache_read":0,"cache_write":0},"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","open_weights":true},
  "glm-4.7-flashx": {"name":"GLM-4.7-FlashX","description":"Efficient GLM model for fast reasoning, coding, and agent workflows","family":"glm-flash","canonical_model_id":"zhipuai/glm-4.7-flashx","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":131072},"cost":{"input":0.07,"output":0.4,"cache_read":0.01,"cache_write":0},"knowledge":"2025-04","release_date":"2026-01-19","last_updated":"2026-01-19","open_weights":true},
  "glm-5": {"name":"GLM-5","description":"General GLM flagship for coding, analysis, and tool-heavy engineering workflows","family":"glm","canonical_model_id":"zhipuai/glm-5","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":204800,"output":131072},"cost":{"input":1,"output":3.2,"cache_read":0.2,"cache_write":0},"release_date":"2026-02-12","last_updated":"2026-02-12","open_weights":true},
  "glm-5-turbo": {"name":"GLM-5-Turbo","description":"Faster GLM-5 lane for coding agents that need lower latency","family":"glm","canonical_model_id":"zhipuai/glm-5-turbo","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24,"cache_write":0},"release_date":"2026-03-16","last_updated":"2026-03-16","open_weights":false},
  "glm-5.1": {"name":"GLM-5.1","description":"Strong GLM coding model for agentic engineering, terminals, and repository generation","family":"glm","canonical_model_id":"zhipuai/glm-5.1","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":200000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26,"cache_write":0},"release_date":"2026-04-07","last_updated":"2026-04-07","open_weights":true},
  "glm-5.2": {"name":"GLM-5.2","description":"Open flagship GLM for long-horizon coding agents and million-token context work","family":"glm","canonical_model_id":"zhipuai/glm-5.2","reasoning":true,"reasoning_options":[{"type":"toggle"},{"type":"effort","values":["high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26,"cache_write":0},"release_date":"2026-06-13","last_updated":"2026-06-13","open_weights":true},
  "glm-5.3": {"name":"GLM-5.3","description":"Flagship GLM model for long-horizon coding, agents, and complex project delivery","family":"glm","canonical_model_id":"zhipuai/glm-5.3","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":false,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":1.4,"output":4.4,"cache_read":0.26,"cache_write":0},"release_date":"2026-08-14","last_updated":"2026-08-14","open_weights":true},
  "glm-5.3-flash": {"name":"GLM-5.3-Flash","description":"Native multimodal GLM model for efficient coding and long-horizon agent tasks","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.15,"output":0.5,"cache_read":0.03,"cache_write":0},"release_date":"2026-08-26","last_updated":"2026-08-26","open_weights":true},
  "glm-5.3-flashx": {"name":"GLM-5.3-FlashX","description":"High-speed GLM-5.3-Flash serving option for coding and agent workflows","family":"glm-flash","canonical_model_id":"zhipuai/glm-5.3-flash","reasoning":true,"reasoning_options":[{"type":"effort","values":["low","high","max"]}],"tool_call":true,"structured_output":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":1000000,"output":131072},"cost":{"input":0.37,"output":1.25,"cache_read":0.075,"cache_write":0},"release_date":"2026-09-18","last_updated":"2026-09-18","open_weights":true},
  "glm-5v-turbo": {"name":"GLM-5V-Turbo","description":"Fast GLM vision model for screenshots, documents, and multimodal agent tasks","family":"glm","canonical_model_id":"zhipuai/glm-5v-turbo","reasoning":true,"reasoning_options":[{"type":"toggle"}],"tool_call":true,"temperature":true,"attachment":true,"interleaved":{"field":"reasoning_content"},"modalities":{"input":["text","image","video","pdf"],"output":["text"]},"limit":{"context":200000,"output":131072},"cost":{"input":1.2,"output":4,"cache_read":0.24,"cache_write":0},"release_date":"2026-04-01","last_updated":"2026-04-01","open_weights":false}
}}
}