#[derive(Clone, Copy, Debug, PartialEq, Eq)]
pub struct ModelCapacity {
pub context_window_tokens: usize,
pub max_completion_tokens: usize,
}
pub const CONSERVATIVE_CONTEXT_WINDOW_TOKENS: usize = 32_768;
pub const CONSERVATIVE_MAX_COMPLETION_TOKENS: usize =
crate::context_budget::DEFAULT_MAX_COMPLETION_TOKENS;
pub fn conservative_model_capacity() -> ModelCapacity {
ModelCapacity {
context_window_tokens: CONSERVATIVE_CONTEXT_WINDOW_TOKENS,
max_completion_tokens: CONSERVATIVE_MAX_COMPLETION_TOKENS,
}
}
pub fn catalog_model_capacity(model_id: &str) -> Option<ModelCapacity> {
let normalized = model_id.trim().to_ascii_lowercase();
let index = LITELLM_MODEL_CAPACITIES
.binary_search_by(|entry| entry.0.cmp(normalized.as_str()))
.ok()?;
let (_, context_window_tokens, max_completion_tokens) = LITELLM_MODEL_CAPACITIES[index];
Some(ModelCapacity {
context_window_tokens,
max_completion_tokens,
})
}
const LITELLM_MODEL_CAPACITIES: &[(&str, usize, usize)] = &[
("ai21.j2-mid-v1", 8191, 8191),
("ai21.j2-ultra-v1", 8191, 8191),
("ai21.jamba-1-5-large-v1:0", 256000, 256000),
("ai21.jamba-1-5-mini-v1:0", 256000, 256000),
("ai21.jamba-instruct-v1:0", 70000, 4096),
("amazon-nova/nova-lite-v1", 300000, 10000),
("amazon-nova/nova-micro-v1", 128000, 10000),
("amazon-nova/nova-premier-v1", 1000000, 10000),
("amazon-nova/nova-pro-v1", 300000, 10000),
("amazon.nova-2-lite-v1:0", 1000000, 64000),
("amazon.nova-2-pro-preview-20251202-v1:0", 1000000, 64000),
("amazon.nova-lite-v1:0", 300000, 10000),
("amazon.nova-micro-v1:0", 128000, 10000),
("amazon.nova-pro-v1:0", 300000, 10000),
("amazon.titan-text-express-v1", 42000, 8000),
("amazon.titan-text-lite-v1", 42000, 4000),
("amazon.titan-text-premier-v1:0", 42000, 32000),
("anthropic.claude-3-5-haiku-20241022-v1:0", 200000, 8192),
("anthropic.claude-3-5-sonnet-20240620-v1:0", 1000000, 4096),
("anthropic.claude-3-5-sonnet-20241022-v2:0", 1000000, 8192),
("anthropic.claude-3-7-sonnet-20240620-v1:0", 200000, 8192),
("anthropic.claude-3-7-sonnet-20250219-v1:0", 200000, 8192),
("anthropic.claude-3-haiku-20240307-v1:0", 200000, 4096),
("anthropic.claude-3-opus-20240229-v1:0", 200000, 4096),
("anthropic.claude-3-sonnet-20240229-v1:0", 200000, 4096),
("anthropic.claude-haiku-4-5-20251001-v1:0", 200000, 64000),
("anthropic.claude-haiku-4-5@20251001", 200000, 64000),
("anthropic.claude-instant-v1", 100000, 8191),
("anthropic.claude-mythos-preview", 1000000, 128000),
("anthropic.claude-opus-4-1-20250805-v1:0", 200000, 32000),
("anthropic.claude-opus-4-20250514-v1:0", 200000, 32000),
("anthropic.claude-opus-4-5-20251101-v1:0", 200000, 64000),
("anthropic.claude-opus-4-6-v1", 1000000, 128000),
("anthropic.claude-opus-4-7", 1000000, 128000),
("anthropic.claude-sonnet-4-20250514-v1:0", 1000000, 64000),
("anthropic.claude-sonnet-4-5-20250929-v1:0", 200000, 64000),
("anthropic.claude-sonnet-4-6", 1000000, 64000),
("anthropic.claude-v1", 100000, 8191),
("anthropic.claude-v2:1", 100000, 8191),
("anyscale/codellama/codellama-34b-instruct-hf", 4096, 4096),
("anyscale/codellama/codellama-70b-instruct-hf", 4096, 4096),
("anyscale/google/gemma-7b-it", 8192, 8192),
("anyscale/huggingfaceh4/zephyr-7b-beta", 16384, 16384),
("anyscale/meta-llama/llama-2-13b-chat-hf", 4096, 4096),
("anyscale/meta-llama/llama-2-70b-chat-hf", 4096, 4096),
("anyscale/meta-llama/llama-2-7b-chat-hf", 4096, 4096),
("anyscale/meta-llama/meta-llama-3-70b-instruct", 8192, 8192),
("anyscale/meta-llama/meta-llama-3-8b-instruct", 8192, 8192),
("anyscale/mistralai/mistral-7b-instruct-v0.1", 16384, 16384),
(
"anyscale/mistralai/mixtral-8x22b-instruct-v0.1",
65536,
65536,
),
(
"anyscale/mistralai/mixtral-8x7b-instruct-v0.1",
16384,
16384,
),
("apac.amazon.nova-2-lite-v1:0", 1000000, 64000),
(
"apac.amazon.nova-2-pro-preview-20251202-v1:0",
1000000,
64000,
),
("apac.amazon.nova-lite-v1:0", 300000, 10000),
("apac.amazon.nova-micro-v1:0", 128000, 10000),
("apac.amazon.nova-pro-v1:0", 300000, 10000),
(
"apac.anthropic.claude-3-5-sonnet-20240620-v1:0",
200000,
4096,
),
(
"apac.anthropic.claude-3-5-sonnet-20241022-v2:0",
200000,
8192,
),
("apac.anthropic.claude-3-haiku-20240307-v1:0", 200000, 4096),
("apac.anthropic.claude-3-sonnet-20240229-v1:0", 200000, 4096),
(
"apac.anthropic.claude-haiku-4-5-20251001-v1:0",
200000,
64000,
),
(
"apac.anthropic.claude-sonnet-4-20250514-v1:0",
1000000,
64000,
),
("au.anthropic.claude-haiku-4-5-20251001-v1:0", 200000, 64000),
("au.anthropic.claude-opus-4-6-v1", 1000000, 128000),
("au.anthropic.claude-opus-4-7", 1000000, 128000),
(
"au.anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
64000,
),
("au.anthropic.claude-sonnet-4-6", 1000000, 64000),
("azure/codex-mini", 200000, 100000),
("azure/command-r-plus", 128000, 4096),
("azure/computer-use-preview", 8192, 1024),
("azure/eu/gpt-4o-2024-08-06", 128000, 16384),
("azure/eu/gpt-4o-2024-11-20", 128000, 16384),
("azure/eu/gpt-4o-mini-2024-07-18", 128000, 16384),
(
"azure/eu/gpt-4o-mini-realtime-preview-2024-12-17",
128000,
4096,
),
("azure/eu/gpt-4o-realtime-preview-2024-10-01", 128000, 4096),
("azure/eu/gpt-4o-realtime-preview-2024-12-17", 128000, 4096),
("azure/eu/gpt-5-2025-08-07", 272000, 128000),
("azure/eu/gpt-5-mini-2025-08-07", 272000, 128000),
("azure/eu/gpt-5-nano-2025-08-07", 272000, 128000),
("azure/eu/gpt-5.1", 272000, 128000),
("azure/eu/gpt-5.1-chat", 128000, 128000),
("azure/eu/gpt-5.1-codex", 272000, 128000),
("azure/eu/gpt-5.1-codex-mini", 272000, 128000),
("azure/eu/o1-2024-12-17", 200000, 100000),
("azure/eu/o1-mini-2024-09-12", 128000, 65536),
("azure/eu/o1-preview-2024-09-12", 128000, 32768),
("azure/eu/o3-mini-2025-01-31", 200000, 100000),
("azure/global-standard/gpt-4o-2024-08-06", 128000, 16384),
("azure/global-standard/gpt-4o-2024-11-20", 128000, 16384),
("azure/global-standard/gpt-4o-mini", 128000, 16384),
("azure/global/gpt-4o-2024-08-06", 128000, 16384),
("azure/global/gpt-4o-2024-11-20", 128000, 16384),
("azure/global/gpt-5.1", 272000, 128000),
("azure/global/gpt-5.1-chat", 128000, 128000),
("azure/global/gpt-5.1-codex", 272000, 128000),
("azure/global/gpt-5.1-codex-mini", 272000, 128000),
("azure/gpt-3.5-turbo", 4097, 4096),
("azure/gpt-3.5-turbo-0125", 16384, 4096),
("azure/gpt-3.5-turbo-instruct-0914", 4097, 4097),
("azure/gpt-35-turbo", 4097, 4096),
("azure/gpt-35-turbo-0125", 16384, 4096),
("azure/gpt-35-turbo-1106", 16384, 4096),
("azure/gpt-35-turbo-16k", 16385, 4096),
("azure/gpt-35-turbo-16k-0613", 16385, 4096),
("azure/gpt-35-turbo-instruct", 4097, 4097),
("azure/gpt-35-turbo-instruct-0914", 4097, 4097),
("azure/gpt-4", 8192, 4096),
("azure/gpt-4-0125-preview", 128000, 4096),
("azure/gpt-4-0613", 8192, 4096),
("azure/gpt-4-1106-preview", 128000, 4096),
("azure/gpt-4-32k", 32768, 4096),
("azure/gpt-4-32k-0613", 32768, 4096),
("azure/gpt-4-turbo", 128000, 4096),
("azure/gpt-4-turbo-2024-04-09", 128000, 4096),
("azure/gpt-4-turbo-vision-preview", 128000, 4096),
("azure/gpt-4.1", 1047576, 32768),
("azure/gpt-4.1-2025-04-14", 1047576, 32768),
("azure/gpt-4.1-mini", 1047576, 32768),
("azure/gpt-4.1-mini-2025-04-14", 1047576, 32768),
("azure/gpt-4.1-nano", 1047576, 32768),
("azure/gpt-4.1-nano-2025-04-14", 1047576, 32768),
("azure/gpt-4.5-preview", 128000, 16384),
("azure/gpt-4o", 128000, 16384),
("azure/gpt-4o-2024-05-13", 128000, 4096),
("azure/gpt-4o-2024-08-06", 128000, 16384),
("azure/gpt-4o-2024-11-20", 128000, 16384),
("azure/gpt-4o-audio-preview-2024-12-17", 128000, 16384),
("azure/gpt-4o-mini", 128000, 16384),
("azure/gpt-4o-mini-2024-07-18", 128000, 16384),
("azure/gpt-4o-mini-audio-preview-2024-12-17", 128000, 16384),
(
"azure/gpt-4o-mini-realtime-preview-2024-12-17",
128000,
4096,
),
("azure/gpt-4o-realtime-preview-2024-10-01", 128000, 4096),
("azure/gpt-4o-realtime-preview-2024-12-17", 128000, 4096),
("azure/gpt-5", 272000, 128000),
("azure/gpt-5-2025-08-07", 272000, 128000),
("azure/gpt-5-chat", 128000, 16384),
("azure/gpt-5-chat-latest", 128000, 16384),
("azure/gpt-5-codex", 272000, 128000),
("azure/gpt-5-mini", 272000, 128000),
("azure/gpt-5-mini-2025-08-07", 272000, 128000),
("azure/gpt-5-nano", 272000, 128000),
("azure/gpt-5-nano-2025-08-07", 272000, 128000),
("azure/gpt-5-pro", 272000, 128000),
("azure/gpt-5.1", 272000, 128000),
("azure/gpt-5.1-2025-11-13", 272000, 128000),
("azure/gpt-5.1-chat", 128000, 128000),
("azure/gpt-5.1-chat-2025-11-13", 128000, 16384),
("azure/gpt-5.1-codex", 272000, 128000),
("azure/gpt-5.1-codex-2025-11-13", 272000, 128000),
("azure/gpt-5.1-codex-max", 272000, 128000),
("azure/gpt-5.1-codex-mini", 272000, 128000),
("azure/gpt-5.1-codex-mini-2025-11-13", 272000, 128000),
("azure/gpt-5.2", 272000, 128000),
("azure/gpt-5.2-2025-12-11", 272000, 128000),
("azure/gpt-5.2-chat", 128000, 16384),
("azure/gpt-5.2-chat-2025-12-11", 128000, 16384),
("azure/gpt-5.2-codex", 272000, 128000),
("azure/gpt-5.2-pro", 272000, 128000),
("azure/gpt-5.2-pro-2025-12-11", 272000, 128000),
("azure/gpt-5.3-chat", 128000, 16384),
("azure/gpt-5.3-codex", 272000, 128000),
("azure/gpt-5.4", 1050000, 128000),
("azure/gpt-5.4-2026-03-05", 1050000, 128000),
("azure/gpt-5.4-mini", 1050000, 128000),
("azure/gpt-5.4-nano", 1050000, 128000),
("azure/gpt-5.4-pro", 1050000, 128000),
("azure/gpt-5.4-pro-2026-03-05", 1050000, 128000),
("azure/gpt-audio-1.5-2026-02-23", 128000, 16384),
("azure/gpt-audio-2025-08-28", 128000, 16384),
("azure/gpt-audio-mini-2025-10-06", 128000, 16384),
("azure/gpt-realtime-1.5-2026-02-23", 32000, 4096),
("azure/gpt-realtime-2025-08-28", 32000, 4096),
("azure/gpt-realtime-mini-2025-10-06", 32000, 4096),
("azure/mistral-large-2402", 32000, 32000),
("azure/mistral-large-latest", 32000, 32000),
("azure/o1", 200000, 100000),
("azure/o1-2024-12-17", 200000, 100000),
("azure/o1-mini", 128000, 65536),
("azure/o1-mini-2024-09-12", 128000, 65536),
("azure/o1-preview", 128000, 32768),
("azure/o1-preview-2024-09-12", 128000, 32768),
("azure/o3", 200000, 100000),
("azure/o3-2025-04-16", 200000, 100000),
("azure/o3-deep-research", 200000, 100000),
("azure/o3-mini", 200000, 100000),
("azure/o3-mini-2025-01-31", 200000, 100000),
("azure/o3-pro", 200000, 100000),
("azure/o3-pro-2025-06-10", 200000, 100000),
("azure/o4-mini", 200000, 100000),
("azure/o4-mini-2025-04-16", 200000, 100000),
("azure/us/gpt-4.1-2025-04-14", 1047576, 32768),
("azure/us/gpt-4.1-mini-2025-04-14", 1047576, 32768),
("azure/us/gpt-4.1-nano-2025-04-14", 1047576, 32768),
("azure/us/gpt-4o-2024-08-06", 128000, 16384),
("azure/us/gpt-4o-2024-11-20", 128000, 16384),
("azure/us/gpt-4o-mini-2024-07-18", 128000, 16384),
(
"azure/us/gpt-4o-mini-realtime-preview-2024-12-17",
128000,
4096,
),
("azure/us/gpt-4o-realtime-preview-2024-10-01", 128000, 4096),
("azure/us/gpt-4o-realtime-preview-2024-12-17", 128000, 4096),
("azure/us/gpt-5-2025-08-07", 272000, 128000),
("azure/us/gpt-5-mini-2025-08-07", 272000, 128000),
("azure/us/gpt-5-nano-2025-08-07", 272000, 128000),
("azure/us/gpt-5.1", 272000, 128000),
("azure/us/gpt-5.1-chat", 128000, 128000),
("azure/us/gpt-5.1-codex", 272000, 128000),
("azure/us/gpt-5.1-codex-mini", 272000, 128000),
("azure/us/o1-2024-12-17", 200000, 100000),
("azure/us/o1-mini-2024-09-12", 128000, 65536),
("azure/us/o1-preview-2024-09-12", 128000, 32768),
("azure/us/o3-2025-04-16", 200000, 100000),
("azure/us/o3-mini-2025-01-31", 200000, 100000),
("azure/us/o4-mini-2025-04-16", 200000, 100000),
("azure_ai/claude-haiku-4-5", 200000, 64000),
("azure_ai/claude-opus-4-1", 200000, 32000),
("azure_ai/claude-opus-4-5", 200000, 64000),
("azure_ai/claude-opus-4-6", 200000, 128000),
("azure_ai/claude-opus-4-7", 200000, 128000),
("azure_ai/claude-sonnet-4-5", 200000, 64000),
("azure_ai/claude-sonnet-4-6", 1000000, 64000),
("azure_ai/deepseek-r1", 128000, 8192),
("azure_ai/deepseek-v3", 128000, 8192),
("azure_ai/deepseek-v3-0324", 128000, 8192),
("azure_ai/deepseek-v3.2", 163840, 163840),
("azure_ai/deepseek-v3.2-speciale", 163840, 163840),
("azure_ai/global/grok-3", 131072, 131072),
("azure_ai/global/grok-3-mini", 131072, 131072),
("azure_ai/gpt-oss-120b", 131072, 131072),
("azure_ai/grok-3", 131072, 131072),
("azure_ai/grok-3-mini", 131072, 131072),
("azure_ai/grok-4", 131072, 131072),
("azure_ai/grok-4-1-fast-non-reasoning", 131072, 131072),
("azure_ai/grok-4-1-fast-reasoning", 131072, 131072),
("azure_ai/grok-4-fast-non-reasoning", 131072, 131072),
("azure_ai/grok-4-fast-reasoning", 131072, 131072),
("azure_ai/grok-code-fast-1", 131072, 131072),
("azure_ai/jais-30b-chat", 8192, 8192),
("azure_ai/jamba-instruct", 70000, 4096),
("azure_ai/kimi-k2.5", 262144, 262144),
("azure_ai/llama-3.2-11b-vision-instruct", 128000, 2048),
("azure_ai/llama-3.2-90b-vision-instruct", 128000, 2048),
("azure_ai/llama-3.3-70b-instruct", 128000, 2048),
(
"azure_ai/llama-4-maverick-17b-128e-instruct-fp8",
1000000,
16384,
),
("azure_ai/llama-4-scout-17b-16e-instruct", 10000000, 16384),
("azure_ai/mai-ds-r1", 128000, 8192),
("azure_ai/meta-llama-3-70b-instruct", 8192, 2048),
("azure_ai/meta-llama-3.1-405b-instruct", 128000, 2048),
("azure_ai/meta-llama-3.1-70b-instruct", 128000, 2048),
("azure_ai/meta-llama-3.1-8b-instruct", 128000, 2048),
("azure_ai/ministral-3b", 128000, 4096),
("azure_ai/mistral-large", 32000, 8191),
("azure_ai/mistral-large-2407", 128000, 4096),
("azure_ai/mistral-large-3", 256000, 8191),
("azure_ai/mistral-large-latest", 128000, 4096),
("azure_ai/mistral-medium-2505", 131072, 8191),
("azure_ai/mistral-nemo", 131072, 4096),
("azure_ai/mistral-small", 32000, 8191),
("azure_ai/mistral-small-2503", 128000, 128000),
("azure_ai/phi-3-medium-128k-instruct", 128000, 4096),
("azure_ai/phi-3-medium-4k-instruct", 4096, 4096),
("azure_ai/phi-3-mini-128k-instruct", 128000, 4096),
("azure_ai/phi-3-mini-4k-instruct", 4096, 4096),
("azure_ai/phi-3-small-128k-instruct", 128000, 4096),
("azure_ai/phi-3-small-8k-instruct", 8192, 4096),
("azure_ai/phi-3.5-mini-instruct", 128000, 4096),
("azure_ai/phi-3.5-moe-instruct", 128000, 4096),
("azure_ai/phi-3.5-vision-instruct", 128000, 4096),
("azure_ai/phi-4", 16384, 16384),
("azure_ai/phi-4-mini-instruct", 131072, 4096),
("azure_ai/phi-4-mini-reasoning", 131072, 4096),
("azure_ai/phi-4-multimodal-instruct", 131072, 4096),
("azure_ai/phi-4-reasoning", 32768, 4096),
("babbage-002", 16384, 4096),
(
"bedrock/*/1-month-commitment/cohere.command-light-text-v14",
4096,
4096,
),
(
"bedrock/*/1-month-commitment/cohere.command-text-v14",
4096,
4096,
),
(
"bedrock/*/6-month-commitment/cohere.command-light-text-v14",
4096,
4096,
),
(
"bedrock/*/6-month-commitment/cohere.command-text-v14",
4096,
4096,
),
(
"bedrock/ap-northeast-1/1-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/ap-northeast-1/1-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/ap-northeast-1/1-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/ap-northeast-1/6-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/ap-northeast-1/6-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/ap-northeast-1/6-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/ap-northeast-1/anthropic.claude-instant-v1",
100000,
8191,
),
("bedrock/ap-northeast-1/anthropic.claude-v1", 100000, 8191),
("bedrock/ap-northeast-1/anthropic.claude-v2:1", 100000, 8191),
("bedrock/ap-northeast-1/deepseek.v3.2", 163840, 163840),
("bedrock/ap-northeast-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/ap-northeast-1/minimax.minimax-m2.5", 1000000, 8192),
(
"bedrock/ap-northeast-1/moonshotai.kimi-k2-thinking",
262144,
262144,
),
(
"bedrock/ap-northeast-1/moonshotai.kimi-k2.5",
262144,
262144,
),
("bedrock/ap-northeast-1/qwen.qwen3-coder-next", 262144, 8192),
("bedrock/ap-south-1/deepseek.v3.2", 163840, 163840),
(
"bedrock/ap-south-1/meta.llama3-70b-instruct-v1:0",
8192,
8192,
),
(
"bedrock/ap-south-1/meta.llama3-8b-instruct-v1:0",
8192,
8192,
),
("bedrock/ap-south-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/ap-south-1/minimax.minimax-m2.5", 1000000, 8192),
(
"bedrock/ap-south-1/moonshotai.kimi-k2-thinking",
262144,
262144,
),
("bedrock/ap-south-1/moonshotai.kimi-k2.5", 262144, 262144),
("bedrock/ap-south-1/qwen.qwen3-coder-next", 262144, 8192),
("bedrock/ap-southeast-2/minimax.minimax-m2.5", 1000000, 8192),
("bedrock/ap-southeast-3/deepseek.v3.2", 163840, 163840),
("bedrock/ap-southeast-3/minimax.minimax-m2.1", 196000, 8192),
("bedrock/ap-southeast-3/minimax.minimax-m2.5", 1000000, 8192),
(
"bedrock/ap-southeast-3/moonshotai.kimi-k2.5",
262144,
262144,
),
("bedrock/ap-southeast-3/qwen.qwen3-coder-next", 262144, 8192),
(
"bedrock/ca-central-1/meta.llama3-70b-instruct-v1:0",
8192,
8192,
),
(
"bedrock/ca-central-1/meta.llama3-8b-instruct-v1:0",
8192,
8192,
),
(
"bedrock/eu-central-1/1-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/eu-central-1/1-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/eu-central-1/1-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/eu-central-1/6-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/eu-central-1/6-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/eu-central-1/6-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/eu-central-1/anthropic.claude-instant-v1",
100000,
8191,
),
("bedrock/eu-central-1/anthropic.claude-v1", 100000, 8191),
("bedrock/eu-central-1/anthropic.claude-v2:1", 100000, 8191),
("bedrock/eu-central-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/eu-central-1/minimax.minimax-m2.5", 1000000, 8192),
("bedrock/eu-central-1/qwen.qwen3-coder-next", 262144, 8192),
("bedrock/eu-north-1/deepseek.v3.2", 163840, 163840),
("bedrock/eu-north-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/eu-north-1/minimax.minimax-m2.5", 1000000, 8192),
("bedrock/eu-north-1/moonshotai.kimi-k2.5", 262144, 262144),
("bedrock/eu-south-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/eu-south-1/minimax.minimax-m2.5", 1000000, 8192),
("bedrock/eu-south-1/qwen.qwen3-coder-next", 262144, 8192),
(
"bedrock/eu-west-1/meta.llama3-70b-instruct-v1:0",
8192,
8192,
),
("bedrock/eu-west-1/meta.llama3-8b-instruct-v1:0", 8192, 8192),
("bedrock/eu-west-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/eu-west-1/minimax.minimax-m2.5", 1000000, 8192),
("bedrock/eu-west-1/qwen.qwen3-coder-next", 262144, 8192),
(
"bedrock/eu-west-2/meta.llama3-70b-instruct-v1:0",
8192,
8192,
),
("bedrock/eu-west-2/meta.llama3-8b-instruct-v1:0", 8192, 8192),
("bedrock/eu-west-2/minimax.minimax-m2.1", 196000, 8192),
("bedrock/eu-west-2/minimax.minimax-m2.5", 1000000, 8192),
("bedrock/eu-west-2/qwen.qwen3-coder-next", 262144, 8192),
(
"bedrock/eu-west-3/mistral.mistral-7b-instruct-v0:2",
32000,
8191,
),
(
"bedrock/eu-west-3/mistral.mistral-large-2402-v1:0",
32000,
8191,
),
(
"bedrock/eu-west-3/mistral.mixtral-8x7b-instruct-v0:1",
32000,
8191,
),
(
"bedrock/invoke/anthropic.claude-3-5-sonnet-20240620-v1:0",
200000,
4096,
),
("bedrock/moonshotai.kimi-k2-thinking", 262144, 262144),
("bedrock/moonshotai.kimi-k2.5", 262144, 262144),
("bedrock/sa-east-1/deepseek.v3.2", 163840, 163840),
(
"bedrock/sa-east-1/meta.llama3-70b-instruct-v1:0",
8192,
8192,
),
("bedrock/sa-east-1/meta.llama3-8b-instruct-v1:0", 8192, 8192),
("bedrock/sa-east-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/sa-east-1/minimax.minimax-m2.5", 1000000, 8192),
(
"bedrock/sa-east-1/moonshotai.kimi-k2-thinking",
262144,
262144,
),
("bedrock/sa-east-1/moonshotai.kimi-k2.5", 262144, 262144),
("bedrock/sa-east-1/qwen.qwen3-coder-next", 262144, 8192),
(
"bedrock/us-east-1/1-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/us-east-1/1-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/us-east-1/1-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/us-east-1/6-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/us-east-1/6-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/us-east-1/6-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/us-east-1/anthropic.claude-instant-v1",
100000,
8191,
),
("bedrock/us-east-1/anthropic.claude-v1", 100000, 8191),
("bedrock/us-east-1/anthropic.claude-v2:1", 100000, 8191),
("bedrock/us-east-1/deepseek.v3.2", 163840, 163840),
(
"bedrock/us-east-1/meta.llama3-70b-instruct-v1:0",
8192,
8192,
),
("bedrock/us-east-1/meta.llama3-8b-instruct-v1:0", 8192, 8192),
("bedrock/us-east-1/minimax.minimax-m2.1", 196000, 8192),
("bedrock/us-east-1/minimax.minimax-m2.5", 1000000, 8192),
(
"bedrock/us-east-1/mistral.mistral-7b-instruct-v0:2",
32000,
8191,
),
(
"bedrock/us-east-1/mistral.mistral-large-2402-v1:0",
32000,
8191,
),
(
"bedrock/us-east-1/mistral.mixtral-8x7b-instruct-v0:1",
32000,
8191,
),
(
"bedrock/us-east-1/moonshotai.kimi-k2-thinking",
262144,
262144,
),
("bedrock/us-east-1/moonshotai.kimi-k2.5", 262144, 262144),
("bedrock/us-east-1/qwen.qwen3-coder-next", 262144, 8192),
("bedrock/us-east-2/deepseek.v3.2", 163840, 163840),
("bedrock/us-east-2/minimax.minimax-m2.1", 196000, 8192),
("bedrock/us-east-2/minimax.minimax-m2.5", 1000000, 8192),
(
"bedrock/us-east-2/moonshotai.kimi-k2-thinking",
262144,
262144,
),
("bedrock/us-east-2/moonshotai.kimi-k2.5", 262144, 262144),
("bedrock/us-east-2/qwen.qwen3-coder-next", 262144, 8192),
("bedrock/us-gov-east-1/amazon.nova-pro-v1:0", 300000, 10000),
(
"bedrock/us-gov-east-1/amazon.titan-text-express-v1",
42000,
8000,
),
(
"bedrock/us-gov-east-1/amazon.titan-text-lite-v1",
42000,
4000,
),
(
"bedrock/us-gov-east-1/amazon.titan-text-premier-v1:0",
42000,
32000,
),
(
"bedrock/us-gov-east-1/anthropic.claude-3-5-sonnet-20240620-v1:0",
200000,
8192,
),
(
"bedrock/us-gov-east-1/anthropic.claude-3-haiku-20240307-v1:0",
200000,
4096,
),
(
"bedrock/us-gov-east-1/anthropic.claude-haiku-4-5-20251001-v1:0",
200000,
64000,
),
(
"bedrock/us-gov-east-1/anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
8192,
),
(
"bedrock/us-gov-east-1/claude-sonnet-4-5-20250929-v1:0",
200000,
8192,
),
(
"bedrock/us-gov-east-1/meta.llama3-70b-instruct-v1:0",
8000,
2048,
),
(
"bedrock/us-gov-east-1/meta.llama3-8b-instruct-v1:0",
8000,
2048,
),
("bedrock/us-gov-west-1/amazon.nova-pro-v1:0", 300000, 10000),
(
"bedrock/us-gov-west-1/amazon.titan-text-express-v1",
42000,
8000,
),
(
"bedrock/us-gov-west-1/amazon.titan-text-lite-v1",
42000,
4000,
),
(
"bedrock/us-gov-west-1/amazon.titan-text-premier-v1:0",
42000,
32000,
),
(
"bedrock/us-gov-west-1/anthropic.claude-3-5-sonnet-20240620-v1:0",
200000,
8192,
),
(
"bedrock/us-gov-west-1/anthropic.claude-3-7-sonnet-20250219-v1:0",
200000,
8192,
),
(
"bedrock/us-gov-west-1/anthropic.claude-3-haiku-20240307-v1:0",
200000,
4096,
),
(
"bedrock/us-gov-west-1/anthropic.claude-haiku-4-5-20251001-v1:0",
200000,
64000,
),
(
"bedrock/us-gov-west-1/anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
8192,
),
(
"bedrock/us-gov-west-1/claude-sonnet-4-5-20250929-v1:0",
200000,
8192,
),
(
"bedrock/us-gov-west-1/meta.llama3-70b-instruct-v1:0",
8000,
2048,
),
(
"bedrock/us-gov-west-1/meta.llama3-8b-instruct-v1:0",
8000,
2048,
),
(
"bedrock/us-west-1/meta.llama3-70b-instruct-v1:0",
8192,
8192,
),
("bedrock/us-west-1/meta.llama3-8b-instruct-v1:0", 8192, 8192),
(
"bedrock/us-west-2/1-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/us-west-2/1-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/us-west-2/1-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/us-west-2/6-month-commitment/anthropic.claude-instant-v1",
100000,
8191,
),
(
"bedrock/us-west-2/6-month-commitment/anthropic.claude-v1",
100000,
8191,
),
(
"bedrock/us-west-2/6-month-commitment/anthropic.claude-v2:1",
100000,
8191,
),
(
"bedrock/us-west-2/anthropic.claude-instant-v1",
100000,
8191,
),
("bedrock/us-west-2/anthropic.claude-v1", 100000, 8191),
("bedrock/us-west-2/anthropic.claude-v2:1", 100000, 8191),
("bedrock/us-west-2/deepseek.v3.2", 163840, 163840),
("bedrock/us-west-2/minimax.minimax-m2.1", 196000, 8192),
("bedrock/us-west-2/minimax.minimax-m2.5", 1000000, 8192),
(
"bedrock/us-west-2/mistral.mistral-7b-instruct-v0:2",
32000,
8191,
),
(
"bedrock/us-west-2/mistral.mistral-large-2402-v1:0",
32000,
8191,
),
(
"bedrock/us-west-2/mistral.mixtral-8x7b-instruct-v0:1",
32000,
8191,
),
(
"bedrock/us-west-2/moonshotai.kimi-k2-thinking",
262144,
262144,
),
("bedrock/us-west-2/moonshotai.kimi-k2.5", 262144, 262144),
("bedrock/us-west-2/qwen.qwen3-coder-next", 262144, 8192),
(
"bedrock/us.anthropic.claude-3-5-haiku-20241022-v1:0",
200000,
8192,
),
("bedrock_mantle/openai.gpt-oss-120b", 131072, 32768),
("bedrock_mantle/openai.gpt-oss-20b", 131072, 32768),
(
"bedrock_mantle/openai.gpt-oss-safeguard-120b",
131072,
65536,
),
("bedrock_mantle/openai.gpt-oss-safeguard-20b", 131072, 65536),
("cerebras/gpt-oss-120b", 131072, 32768),
("cerebras/llama-3.3-70b", 128000, 128000),
("cerebras/llama3.1-70b", 128000, 128000),
("cerebras/llama3.1-8b", 128000, 128000),
("cerebras/qwen-3-32b", 128000, 128000),
("cerebras/zai-glm-4.6", 128000, 128000),
("cerebras/zai-glm-4.7", 128000, 128000),
("chatdolphin", 16384, 16384),
("chatgpt-4o-latest", 128000, 4096),
("chatgpt/gpt-5.1-codex-max", 128000, 128000),
("chatgpt/gpt-5.1-codex-mini", 128000, 64000),
("chatgpt/gpt-5.2", 128000, 64000),
("chatgpt/gpt-5.2-codex", 128000, 128000),
("chatgpt/gpt-5.3-chat-latest", 128000, 64000),
("chatgpt/gpt-5.3-codex", 128000, 128000),
("chatgpt/gpt-5.3-codex-spark", 128000, 128000),
("chatgpt/gpt-5.3-instant", 128000, 64000),
("chatgpt/gpt-5.4", 1050000, 128000),
("chatgpt/gpt-5.4-pro", 1050000, 128000),
("claude-3-7-sonnet-20250219", 200000, 64000),
("claude-3-haiku-20240307", 200000, 4096),
("claude-3-opus-20240229", 200000, 4096),
("claude-4-opus-20250514", 200000, 32000),
("claude-4-sonnet-20250514", 1000000, 64000),
("claude-haiku-4-5", 200000, 64000),
("claude-haiku-4-5-20251001", 200000, 64000),
("claude-opus-4-1", 200000, 32000),
("claude-opus-4-1-20250805", 200000, 32000),
("claude-opus-4-20250514", 200000, 32000),
("claude-opus-4-5", 200000, 64000),
("claude-opus-4-5-20251101", 200000, 64000),
("claude-opus-4-6", 1000000, 128000),
("claude-opus-4-6-20260205", 1000000, 128000),
("claude-opus-4-7", 1000000, 128000),
("claude-opus-4-7-20260416", 1000000, 128000),
("claude-sonnet-4-20250514", 1000000, 64000),
("claude-sonnet-4-5", 200000, 64000),
("claude-sonnet-4-5-20250929", 200000, 64000),
("claude-sonnet-4-5-20250929-v1:0", 200000, 64000),
("claude-sonnet-4-6", 1000000, 64000),
("cloudflare/@cf/meta/llama-2-7b-chat-fp16", 3072, 3072),
("cloudflare/@cf/meta/llama-2-7b-chat-int8", 2048, 2048),
(
"cloudflare/@cf/mistral/mistral-7b-instruct-v0.1",
8192,
8192,
),
(
"cloudflare/@hf/thebloke/codellama-7b-instruct-awq",
4096,
4096,
),
("codestral/codestral-2405", 32000, 8191),
("codestral/codestral-latest", 32000, 8191),
("codex-mini-latest", 200000, 100000),
("cohere.command-light-text-v14", 4096, 4096),
("cohere.command-r-plus-v1:0", 128000, 4096),
("cohere.command-r-v1:0", 128000, 4096),
("cohere.command-text-v14", 4096, 4096),
("command", 4096, 4096),
("command-a-03-2025", 256000, 8000),
("command-light", 4096, 4096),
("command-nightly", 4096, 4096),
("command-r", 128000, 4096),
("command-r-08-2024", 128000, 4096),
("command-r-plus", 128000, 4096),
("command-r-plus-08-2024", 128000, 4096),
("command-r7b-12-2024", 128000, 4096),
("computer-use-preview", 8192, 1024),
("dashscope/qwen-coder", 1000000, 16384),
("dashscope/qwen-flash", 997952, 32768),
("dashscope/qwen-flash-2025-07-28", 997952, 32768),
("dashscope/qwen-max", 30720, 8192),
("dashscope/qwen-plus", 129024, 16384),
("dashscope/qwen-plus-2025-01-25", 129024, 8192),
("dashscope/qwen-plus-2025-04-28", 129024, 16384),
("dashscope/qwen-plus-2025-07-14", 129024, 16384),
("dashscope/qwen-plus-2025-07-28", 997952, 32768),
("dashscope/qwen-plus-2025-09-11", 997952, 32768),
("dashscope/qwen-plus-latest", 997952, 32768),
("dashscope/qwen-turbo", 129024, 16384),
("dashscope/qwen-turbo-2024-11-01", 1000000, 8192),
("dashscope/qwen-turbo-2025-04-28", 1000000, 16384),
("dashscope/qwen-turbo-latest", 1000000, 16384),
("dashscope/qwen3-30b-a3b", 129024, 16384),
("dashscope/qwen3-coder-flash", 997952, 65536),
("dashscope/qwen3-coder-flash-2025-07-28", 997952, 65536),
("dashscope/qwen3-coder-plus", 997952, 65536),
("dashscope/qwen3-coder-plus-2025-07-22", 997952, 65536),
("dashscope/qwen3-max", 258048, 65536),
("dashscope/qwen3-max-2026-01-23", 258048, 65536),
("dashscope/qwen3-max-preview", 258048, 65536),
("dashscope/qwen3-next-80b-a3b-instruct", 262144, 65536),
("dashscope/qwen3-next-80b-a3b-thinking", 262144, 65536),
("dashscope/qwen3-vl-235b-a22b-instruct", 131072, 32768),
("dashscope/qwen3-vl-235b-a22b-thinking", 131072, 32768),
("dashscope/qwen3-vl-32b-instruct", 131072, 32768),
("dashscope/qwen3-vl-32b-thinking", 131072, 32768),
("dashscope/qwen3-vl-plus", 260096, 32768),
("dashscope/qwen3.5-plus", 991808, 65536),
("dashscope/qwq-plus", 98304, 8192),
("databricks/databricks-claude-3-7-sonnet", 200000, 128000),
("databricks/databricks-claude-haiku-4-5", 200000, 64000),
("databricks/databricks-claude-opus-4", 200000, 32000),
("databricks/databricks-claude-opus-4-1", 200000, 32000),
("databricks/databricks-claude-opus-4-5", 200000, 64000),
("databricks/databricks-claude-sonnet-4", 200000, 64000),
("databricks/databricks-claude-sonnet-4-1", 200000, 64000),
("databricks/databricks-claude-sonnet-4-5", 200000, 64000),
("databricks/databricks-gemini-2-5-flash", 1048576, 65535),
("databricks/databricks-gemini-2-5-pro", 1048576, 65536),
("databricks/databricks-gemma-3-12b", 128000, 32000),
("databricks/databricks-gpt-5", 272000, 128000),
("databricks/databricks-gpt-5-1", 272000, 128000),
("databricks/databricks-gpt-5-mini", 272000, 128000),
("databricks/databricks-gpt-5-nano", 272000, 128000),
("databricks/databricks-gpt-oss-120b", 131072, 131072),
("databricks/databricks-gpt-oss-20b", 131072, 131072),
("databricks/databricks-llama-2-70b-chat", 4096, 4096),
("databricks/databricks-llama-4-maverick", 128000, 128000),
(
"databricks/databricks-meta-llama-3-1-405b-instruct",
128000,
128000,
),
(
"databricks/databricks-meta-llama-3-1-8b-instruct",
200000,
128000,
),
(
"databricks/databricks-meta-llama-3-3-70b-instruct",
128000,
128000,
),
(
"databricks/databricks-meta-llama-3-70b-instruct",
128000,
128000,
),
("databricks/databricks-mixtral-8x7b-instruct", 4096, 4096),
("databricks/databricks-mpt-30b-instruct", 8192, 8192),
("databricks/databricks-mpt-7b-instruct", 8192, 8192),
("davinci-002", 16384, 4096),
("deepinfra/allenai/olmocr-7b-0725-fp8", 16384, 16384),
(
"deepinfra/anthropic/claude-3-7-sonnet-latest",
200000,
200000,
),
("deepinfra/anthropic/claude-4-opus", 200000, 200000),
("deepinfra/anthropic/claude-4-sonnet", 200000, 200000),
("deepinfra/deepseek-ai/deepseek-r1", 163840, 163840),
("deepinfra/deepseek-ai/deepseek-r1-0528", 163840, 163840),
("deepinfra/deepseek-ai/deepseek-r1-0528-turbo", 32768, 32768),
(
"deepinfra/deepseek-ai/deepseek-r1-distill-llama-70b",
131072,
131072,
),
(
"deepinfra/deepseek-ai/deepseek-r1-distill-qwen-32b",
131072,
131072,
),
("deepinfra/deepseek-ai/deepseek-r1-turbo", 40960, 40960),
("deepinfra/deepseek-ai/deepseek-v3", 163840, 163840),
("deepinfra/deepseek-ai/deepseek-v3-0324", 163840, 163840),
("deepinfra/deepseek-ai/deepseek-v3.1", 163840, 163840),
(
"deepinfra/deepseek-ai/deepseek-v3.1-terminus",
163840,
163840,
),
("deepinfra/google/gemini-2.0-flash-001", 1000000, 1000000),
("deepinfra/google/gemini-2.5-flash", 1000000, 1000000),
("deepinfra/google/gemini-2.5-pro", 1000000, 1000000),
("deepinfra/google/gemma-3-12b-it", 131072, 131072),
("deepinfra/google/gemma-3-27b-it", 131072, 131072),
("deepinfra/google/gemma-3-4b-it", 131072, 131072),
("deepinfra/gryphe/mythomax-l2-13b", 4096, 4096),
(
"deepinfra/meta-llama/llama-3.2-11b-vision-instruct",
131072,
131072,
),
("deepinfra/meta-llama/llama-3.2-3b-instruct", 131072, 131072),
(
"deepinfra/meta-llama/llama-3.3-70b-instruct",
131072,
131072,
),
(
"deepinfra/meta-llama/llama-3.3-70b-instruct-turbo",
131072,
131072,
),
(
"deepinfra/meta-llama/llama-4-maverick-17b-128e-instruct-fp8",
1048576,
1048576,
),
(
"deepinfra/meta-llama/llama-4-scout-17b-16e-instruct",
327680,
327680,
),
("deepinfra/meta-llama/llama-guard-3-8b", 131072, 131072),
("deepinfra/meta-llama/llama-guard-4-12b", 163840, 163840),
("deepinfra/meta-llama/meta-llama-3-8b-instruct", 8192, 8192),
(
"deepinfra/meta-llama/meta-llama-3.1-70b-instruct",
131072,
131072,
),
(
"deepinfra/meta-llama/meta-llama-3.1-70b-instruct-turbo",
131072,
131072,
),
(
"deepinfra/meta-llama/meta-llama-3.1-8b-instruct",
131072,
131072,
),
(
"deepinfra/meta-llama/meta-llama-3.1-8b-instruct-turbo",
131072,
131072,
),
("deepinfra/microsoft/phi-4", 16384, 16384),
("deepinfra/microsoft/wizardlm-2-8x22b", 65536, 65536),
(
"deepinfra/mistralai/mistral-nemo-instruct-2407",
131072,
131072,
),
(
"deepinfra/mistralai/mistral-small-24b-instruct-2501",
32768,
32768,
),
(
"deepinfra/mistralai/mistral-small-3.2-24b-instruct-2506",
128000,
128000,
),
(
"deepinfra/mistralai/mixtral-8x7b-instruct-v0.1",
32768,
32768,
),
("deepinfra/moonshotai/kimi-k2-instruct", 131072, 131072),
("deepinfra/moonshotai/kimi-k2-instruct-0905", 262144, 262144),
(
"deepinfra/nousresearch/hermes-3-llama-3.1-405b",
131072,
131072,
),
(
"deepinfra/nousresearch/hermes-3-llama-3.1-70b",
131072,
131072,
),
(
"deepinfra/nvidia/llama-3.1-nemotron-70b-instruct",
131072,
131072,
),
(
"deepinfra/nvidia/llama-3.3-nemotron-super-49b-v1.5",
131072,
131072,
),
(
"deepinfra/nvidia/nvidia-nemotron-nano-9b-v2",
131072,
131072,
),
("deepinfra/openai/gpt-oss-120b", 131072, 131072),
("deepinfra/openai/gpt-oss-20b", 131072, 131072),
("deepinfra/qwen/qwen2.5-72b-instruct", 32768, 32768),
("deepinfra/qwen/qwen2.5-7b-instruct", 32768, 32768),
("deepinfra/qwen/qwen2.5-vl-32b-instruct", 128000, 128000),
("deepinfra/qwen/qwen3-14b", 40960, 40960),
("deepinfra/qwen/qwen3-235b-a22b", 40960, 40960),
(
"deepinfra/qwen/qwen3-235b-a22b-instruct-2507",
262144,
262144,
),
(
"deepinfra/qwen/qwen3-235b-a22b-thinking-2507",
262144,
262144,
),
("deepinfra/qwen/qwen3-30b-a3b", 40960, 40960),
("deepinfra/qwen/qwen3-32b", 40960, 40960),
(
"deepinfra/qwen/qwen3-coder-480b-a35b-instruct",
262144,
262144,
),
(
"deepinfra/qwen/qwen3-coder-480b-a35b-instruct-turbo",
262144,
262144,
),
("deepinfra/qwen/qwen3-next-80b-a3b-instruct", 262144, 262144),
("deepinfra/qwen/qwen3-next-80b-a3b-thinking", 262144, 262144),
("deepinfra/qwen/qwq-32b", 131072, 131072),
("deepinfra/sao10k/l3-8b-lunaris-v1-turbo", 8192, 8192),
("deepinfra/sao10k/l3.1-70b-euryale-v2.2", 131072, 131072),
("deepinfra/sao10k/l3.3-70b-euryale-v2.3", 131072, 131072),
("deepinfra/zai-org/glm-4.5", 131072, 131072),
("deepseek-chat", 1048576, 393216),
("deepseek-reasoner", 1048576, 393216),
("deepseek-v3-2-251201", 98304, 32768),
("deepseek-v4-flash", 1048576, 393216),
("deepseek-v4-pro", 1048576, 393216),
("deepseek.v3-v1:0", 163840, 81920),
("deepseek.v3.2", 163840, 163840),
("deepseek/deepseek-chat", 1048576, 393216),
("deepseek/deepseek-coder", 128000, 4096),
("deepseek/deepseek-r1", 65536, 8192),
("deepseek/deepseek-reasoner", 1048576, 393216),
("deepseek/deepseek-v3", 65536, 8192),
("deepseek/deepseek-v3.2", 163840, 163840),
("deepseek/deepseek-v4-flash", 1048576, 393216),
("deepseek/deepseek-v4-pro", 1048576, 393216),
("dolphin", 16384, 16384),
("eu.amazon.nova-2-lite-v1:0", 1000000, 64000),
("eu.amazon.nova-2-pro-preview-20251202-v1:0", 1000000, 64000),
("eu.amazon.nova-lite-v1:0", 300000, 10000),
("eu.amazon.nova-micro-v1:0", 128000, 10000),
("eu.amazon.nova-pro-v1:0", 300000, 10000),
("eu.anthropic.claude-3-5-haiku-20241022-v1:0", 200000, 8192),
("eu.anthropic.claude-3-5-sonnet-20240620-v1:0", 200000, 4096),
("eu.anthropic.claude-3-5-sonnet-20241022-v2:0", 200000, 8192),
("eu.anthropic.claude-3-7-sonnet-20250219-v1:0", 200000, 8192),
("eu.anthropic.claude-3-haiku-20240307-v1:0", 200000, 4096),
("eu.anthropic.claude-3-opus-20240229-v1:0", 200000, 4096),
("eu.anthropic.claude-3-sonnet-20240229-v1:0", 200000, 4096),
("eu.anthropic.claude-haiku-4-5-20251001-v1:0", 200000, 64000),
("eu.anthropic.claude-opus-4-1-20250805-v1:0", 200000, 32000),
("eu.anthropic.claude-opus-4-20250514-v1:0", 200000, 32000),
("eu.anthropic.claude-opus-4-5-20251101-v1:0", 200000, 64000),
("eu.anthropic.claude-opus-4-6-v1", 1000000, 128000),
("eu.anthropic.claude-opus-4-7", 1000000, 128000),
("eu.anthropic.claude-sonnet-4-20250514-v1:0", 1000000, 64000),
(
"eu.anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
64000,
),
("eu.anthropic.claude-sonnet-4-6", 1000000, 64000),
("eu.deepseek.v3.2", 163840, 163840),
("eu.meta.llama3-2-1b-instruct-v1:0", 128000, 4096),
("eu.meta.llama3-2-3b-instruct-v1:0", 128000, 4096),
("eu.mistral.pixtral-large-2502-v1:0", 128000, 4096),
("featherless_ai/featherless-ai/qwerky-72b", 32768, 4096),
("featherless_ai/featherless-ai/qwerky-qwq-32b", 32768, 4096),
(
"fireworks_ai/accounts/fireworks/models/chronos-hermes-13b-v2",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-13b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-13b-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-13b-python",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-34b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-34b-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-34b-python",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-70b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-70b-instruct",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-70b-python",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-7b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-7b-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-llama-7b-python",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/code-qwen-1p5-7b",
65536,
65536,
),
(
"fireworks_ai/accounts/fireworks/models/codegemma-2b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/codegemma-7b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/cogito-671b-v2-p1",
163840,
163840,
),
(
"fireworks_ai/accounts/fireworks/models/cogito-v1-preview-llama-3b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/cogito-v1-preview-llama-70b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/cogito-v1-preview-llama-8b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/cogito-v1-preview-qwen-14b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/cogito-v1-preview-qwen-32b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/dbrx-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-1b-base",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-33b-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-7b-base",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-7b-base-v1p5",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-7b-instruct-v1p5",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-v2-instruct",
65536,
65536,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-v2-lite-base",
163840,
163840,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-coder-v2-lite-instruct",
163840,
163840,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-prover-v2",
163840,
163840,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1",
128000,
20480,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-0528",
160000,
160000,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-0528-distill-qwen3-8b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-basic",
128000,
20480,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-distill-llama-70b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-distill-llama-8b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-distill-qwen-14b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-distill-qwen-1p5b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-distill-qwen-32b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-r1-distill-qwen-7b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-v2-lite-chat",
163840,
163840,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-v2p5",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-v3",
128000,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-v3-0324",
163840,
163840,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-v3p1",
128000,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-v3p1-terminus",
128000,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/deepseek-v3p2",
163840,
163840,
),
(
"fireworks_ai/accounts/fireworks/models/devstral-small-2505",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/dobby-mini-unhinged-plus-llama-3-1-8b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/dobby-unhinged-llama-3-3-70b-new",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/dolphin-2-9-2-qwen2-72b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/dolphin-2p6-mixtral-8x7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/ernie-4p5-21b-a3b-pt",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/ernie-4p5-300b-a47b-pt",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/fare-20b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/firefunction-v1",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/firefunction-v2",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/firellava-13b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/firesearch-ocr-v6",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/flux-1-dev",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/flux-1-dev-controlnet-union",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/flux-1-schnell",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/gemma-2b-it",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/gemma-3-27b-it",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/gemma-7b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/gemma-7b-it",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/gemma2-9b-it",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/glm-4p5",
128000,
96000,
),
(
"fireworks_ai/accounts/fireworks/models/glm-4p5-air",
128000,
96000,
),
(
"fireworks_ai/accounts/fireworks/models/glm-4p5v",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/glm-4p6",
202800,
202800,
),
(
"fireworks_ai/accounts/fireworks/models/glm-4p7",
202800,
202800,
),
(
"fireworks_ai/accounts/fireworks/models/gpt-oss-120b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/gpt-oss-20b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/gpt-oss-safeguard-120b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/gpt-oss-safeguard-20b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/hermes-2-pro-mistral-7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/internvl3-38b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/internvl3-78b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/internvl3-8b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/kat-coder",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/kat-dev-32b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/kat-dev-72b-exp",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/kimi-k2-instruct",
131072,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/kimi-k2-instruct-0905",
262144,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/kimi-k2-thinking",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/kimi-k2p5",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/llama-guard-2-8b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/llama-guard-3-1b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama-guard-3-8b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v2-13b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v2-13b-chat",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v2-70b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v2-70b-chat",
2048,
2048,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v2-7b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v2-7b-chat",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3-70b-instruct",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3-70b-instruct-hf",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3-8b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3-8b-instruct-hf",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p1-405b-instruct",
128000,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p1-405b-instruct-long",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p1-70b-instruct",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p1-70b-instruct-1b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p1-8b-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p1-nemotron-70b-instruct",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p2-11b-vision-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p2-1b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p2-1b-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p2-3b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p2-3b-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p2-90b-vision-instruct",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/llama-v3p3-70b-instruct",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama4-maverick-instruct-basic",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llama4-scout-instruct-basic",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/llamaguard-7b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/llava-yi-34b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/minimax-m1-80k",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/minimax-m2",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/minimax-m2p1",
204800,
204800,
),
(
"fireworks_ai/accounts/fireworks/models/ministral-3-14b-instruct-2512",
256000,
256000,
),
(
"fireworks_ai/accounts/fireworks/models/ministral-3-3b-instruct-2512",
256000,
256000,
),
(
"fireworks_ai/accounts/fireworks/models/ministral-3-8b-instruct-2512",
256000,
256000,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-7b-instruct-4k",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-7b-instruct-v0p2",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-7b-instruct-v3",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-7b-v0p2",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-large-3-fp8",
256000,
256000,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-nemo-base-2407",
128000,
128000,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-nemo-instruct-2407",
128000,
128000,
),
(
"fireworks_ai/accounts/fireworks/models/mistral-small-24b-instruct-2501",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mixtral-8x22b",
65536,
65536,
),
(
"fireworks_ai/accounts/fireworks/models/mixtral-8x22b-instruct",
65536,
65536,
),
(
"fireworks_ai/accounts/fireworks/models/mixtral-8x22b-instruct-hf",
65536,
65536,
),
(
"fireworks_ai/accounts/fireworks/models/mixtral-8x7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mixtral-8x7b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mixtral-8x7b-instruct-hf",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/mythomax-l2-13b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/nemotron-nano-v2-12b-vl",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/nous-capybara-7b-v1p9",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/nous-hermes-2-mixtral-8x7b-dpo",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/nous-hermes-2-yi-34b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/nous-hermes-llama2-13b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/nous-hermes-llama2-70b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/nous-hermes-llama2-7b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/nvidia-nemotron-nano-12b-v2",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/nvidia-nemotron-nano-9b-v2",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/openchat-3p5-0106-7b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/openhermes-2-mistral-7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/openhermes-2p5-mistral-7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/openorca-7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/phi-2-3b",
2048,
2048,
),
(
"fireworks_ai/accounts/fireworks/models/phi-3-mini-128k-instruct",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/phi-3-vision-128k-instruct",
32064,
32064,
),
(
"fireworks_ai/accounts/fireworks/models/phind-code-llama-34b-python-v1",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/phind-code-llama-34b-v1",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/phind-code-llama-34b-v2",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/pythia-12b",
2048,
2048,
),
(
"fireworks_ai/accounts/fireworks/models/qwen-qwq-32b-preview",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen-v2p5-14b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen-v2p5-7b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen1p5-72b-chat",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2-72b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2-7b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2-vl-2b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2-vl-72b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2-vl-7b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-0p5b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-14b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-1p5b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-32b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-32b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-72b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-72b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-7b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-0p5b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-0p5b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-14b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-14b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-1p5b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-1p5b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-32b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-32b-instruct",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-32b-instruct-128k",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-32b-instruct-32k-rope",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-32b-instruct-64k",
65536,
65536,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-3b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-3b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-7b",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-coder-7b-instruct",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-math-72b-instruct",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-vl-32b-instruct",
128000,
128000,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-vl-3b-instruct",
128000,
128000,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-vl-72b-instruct",
128000,
128000,
),
(
"fireworks_ai/accounts/fireworks/models/qwen2p5-vl-7b-instruct",
128000,
128000,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-0p6b",
40960,
40960,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-14b",
40960,
40960,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-1p7b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-1p7b-fp8-draft",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-1p7b-fp8-draft-131072",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-1p7b-fp8-draft-40960",
40960,
40960,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-235b-a22b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-235b-a22b-instruct-2507",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-235b-a22b-thinking-2507",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-30b-a3b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-30b-a3b-instruct-2507",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-30b-a3b-thinking-2507",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-32b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-4b",
40960,
40960,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-4b-instruct-2507",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-8b",
40960,
40960,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-coder-30b-a3b-instruct",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-coder-480b-a35b-instruct",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-coder-480b-instruct-bf16",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-next-80b-a3b-instruct",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-next-80b-a3b-thinking",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-vl-235b-a22b-instruct",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-vl-235b-a22b-thinking",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-vl-30b-a3b-instruct",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-vl-30b-a3b-thinking",
262144,
262144,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-vl-32b-instruct",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/qwen3-vl-8b-instruct",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/qwq-32b",
131072,
131072,
),
(
"fireworks_ai/accounts/fireworks/models/rolm-ocr",
128000,
128000,
),
(
"fireworks_ai/accounts/fireworks/models/snorkel-mistral-7b-pairrm-dpo",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/stablecode-3b",
4096,
4096,
),
(
"fireworks_ai/accounts/fireworks/models/starcoder-16b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/starcoder-7b",
8192,
8192,
),
(
"fireworks_ai/accounts/fireworks/models/starcoder2-15b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/starcoder2-3b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/starcoder2-7b",
16384,
16384,
),
(
"fireworks_ai/accounts/fireworks/models/toppy-m-7b",
32768,
32768,
),
("fireworks_ai/accounts/fireworks/models/yi-34b", 4096, 4096),
(
"fireworks_ai/accounts/fireworks/models/yi-34b-200k-capybara",
200000,
200000,
),
(
"fireworks_ai/accounts/fireworks/models/yi-34b-chat",
4096,
4096,
),
("fireworks_ai/accounts/fireworks/models/yi-6b", 4096, 4096),
(
"fireworks_ai/accounts/fireworks/models/yi-large",
32768,
32768,
),
(
"fireworks_ai/accounts/fireworks/models/zephyr-7b-beta",
32768,
32768,
),
("fireworks_ai/glm-4p7", 202800, 202800),
("fireworks_ai/kimi-k2p5", 262144, 262144),
("fireworks_ai/minimax-m2p1", 204800, 204800),
("friendliai/meta-llama-3.1-70b-instruct", 8192, 8192),
("friendliai/meta-llama-3.1-8b-instruct", 8192, 8192),
("ft:babbage-002", 16384, 4096),
("ft:davinci-002", 16384, 4096),
("ft:gpt-3.5-turbo", 16385, 4096),
("ft:gpt-3.5-turbo-0125", 16385, 4096),
("ft:gpt-3.5-turbo-0613", 4096, 4096),
("ft:gpt-3.5-turbo-1106", 16385, 4096),
("ft:gpt-4-0613", 8192, 4096),
("ft:gpt-4.1-2025-04-14", 1047576, 32768),
("ft:gpt-4.1-mini-2025-04-14", 1047576, 32768),
("ft:gpt-4.1-nano-2025-04-14", 1047576, 32768),
("ft:gpt-4o-2024-08-06", 128000, 16384),
("ft:gpt-4o-2024-11-20", 128000, 16384),
("ft:gpt-4o-mini-2024-07-18", 128000, 16384),
("ft:o4-mini-2025-04-16", 200000, 100000),
("gemini-2.0-flash", 1048576, 8192),
("gemini-2.0-flash-001", 1048576, 8192),
("gemini-2.0-flash-lite", 1048576, 8192),
("gemini-2.0-flash-lite-001", 1048576, 8192),
("gemini-2.5-computer-use-preview-10-2025", 128000, 64000),
("gemini-2.5-flash", 1048576, 65535),
("gemini-2.5-flash-lite", 1048576, 65535),
("gemini-2.5-flash-lite-preview-06-17", 1048576, 65535),
("gemini-2.5-flash-lite-preview-09-2025", 1048576, 65535),
("gemini-2.5-flash-native-audio-latest", 1048576, 8192),
(
"gemini-2.5-flash-native-audio-preview-09-2025",
1048576,
8192,
),
(
"gemini-2.5-flash-native-audio-preview-12-2025",
1048576,
8192,
),
("gemini-2.5-flash-preview-09-2025", 1048576, 65535),
("gemini-2.5-pro", 1048576, 65535),
("gemini-2.5-pro-preview-tts", 1048576, 65535),
("gemini-3-flash-preview", 1048576, 65535),
("gemini-3-pro-preview", 1048576, 65535),
("gemini-3.1-flash-lite-preview", 1048576, 65536),
("gemini-3.1-flash-live-preview", 131072, 65536),
("gemini-3.1-pro-preview", 1048576, 65536),
("gemini-3.1-pro-preview-customtools", 1048576, 65536),
("gemini-exp-1206", 1048576, 65535),
("gemini-flash-latest", 1048576, 65535),
("gemini-flash-lite-latest", 1048576, 65535),
("gemini-pro-latest", 1048576, 65535),
("gemini-robotics-er-1.5-preview", 1048576, 65535),
("gemini/gemini-2.0-flash", 1048576, 8192),
("gemini/gemini-2.0-flash-001", 1048576, 8192),
("gemini/gemini-2.0-flash-lite", 1048576, 8192),
("gemini/gemini-2.0-flash-lite-001", 1048576, 8192),
(
"gemini/gemini-2.5-computer-use-preview-10-2025",
128000,
64000,
),
("gemini/gemini-2.5-flash", 1048576, 65535),
("gemini/gemini-2.5-flash-lite", 1048576, 65535),
("gemini/gemini-2.5-flash-lite-preview-06-17", 1048576, 65535),
(
"gemini/gemini-2.5-flash-lite-preview-09-2025",
1048576,
65535,
),
("gemini/gemini-2.5-flash-native-audio-latest", 1048576, 8192),
(
"gemini/gemini-2.5-flash-native-audio-preview-09-2025",
1048576,
8192,
),
(
"gemini/gemini-2.5-flash-native-audio-preview-12-2025",
1048576,
8192,
),
("gemini/gemini-2.5-flash-preview-09-2025", 1048576, 65535),
("gemini/gemini-2.5-pro", 1048576, 65535),
("gemini/gemini-2.5-pro-preview-tts", 1048576, 65535),
("gemini/gemini-3-flash-preview", 1048576, 65535),
("gemini/gemini-3-pro-preview", 1048576, 65535),
("gemini/gemini-3.1-flash-lite-preview", 1048576, 65536),
("gemini/gemini-3.1-flash-live-preview", 131072, 65536),
("gemini/gemini-3.1-pro-preview", 1048576, 65536),
("gemini/gemini-3.1-pro-preview-customtools", 1048576, 65536),
("gemini/gemini-exp-1114", 1048576, 8192),
("gemini/gemini-exp-1206", 2097152, 8192),
("gemini/gemini-flash-latest", 1048576, 65535),
("gemini/gemini-flash-lite-latest", 1048576, 65535),
("gemini/gemini-gemma-2-27b-it", 8192, 8192),
("gemini/gemini-gemma-2-9b-it", 8192, 8192),
("gemini/gemini-pro-latest", 1048576, 65535),
("gemini/gemini-robotics-er-1.5-preview", 1048576, 65535),
("gemini/gemma-3-27b-it", 131072, 8192),
("gemini/learnlm-1.5-pro-experimental", 32767, 8192),
("gemini/lyria-3-clip-preview", 131072, 8192),
("gemini/lyria-3-pro-preview", 131072, 8192),
("gigachat/gigachat-2-lite", 128000, 8192),
("gigachat/gigachat-2-max", 128000, 8192),
("gigachat/gigachat-2-pro", 128000, 8192),
("github_copilot/claude-haiku-4.5", 128000, 16000),
("github_copilot/claude-opus-4.5", 128000, 16000),
("github_copilot/claude-opus-4.6-fast", 128000, 16000),
("github_copilot/claude-opus-41", 80000, 16000),
("github_copilot/claude-sonnet-4", 128000, 16000),
("github_copilot/claude-sonnet-4.5", 128000, 16000),
("github_copilot/gemini-2.5-pro", 128000, 64000),
("github_copilot/gemini-3-pro-preview", 128000, 64000),
("github_copilot/gpt-3.5-turbo", 16384, 4096),
("github_copilot/gpt-3.5-turbo-0613", 16384, 4096),
("github_copilot/gpt-4", 32768, 4096),
("github_copilot/gpt-4-0613", 32768, 4096),
("github_copilot/gpt-4-o-preview", 64000, 4096),
("github_copilot/gpt-4.1", 128000, 16384),
("github_copilot/gpt-4.1-2025-04-14", 128000, 16384),
("github_copilot/gpt-4o", 64000, 4096),
("github_copilot/gpt-4o-2024-05-13", 64000, 4096),
("github_copilot/gpt-4o-2024-08-06", 64000, 16384),
("github_copilot/gpt-4o-2024-11-20", 64000, 16384),
("github_copilot/gpt-4o-mini", 64000, 4096),
("github_copilot/gpt-4o-mini-2024-07-18", 64000, 4096),
("github_copilot/gpt-5", 128000, 128000),
("github_copilot/gpt-5-mini", 128000, 64000),
("github_copilot/gpt-5.1", 128000, 64000),
("github_copilot/gpt-5.1-codex-max", 128000, 128000),
("github_copilot/gpt-5.2", 128000, 64000),
("github_copilot/gpt-5.3-codex", 128000, 128000),
("glm-4-7-251222", 204800, 131072),
("global.amazon.nova-2-lite-v1:0", 1000000, 64000),
(
"global.anthropic.claude-haiku-4-5-20251001-v1:0",
200000,
64000,
),
(
"global.anthropic.claude-opus-4-5-20251101-v1:0",
200000,
64000,
),
("global.anthropic.claude-opus-4-6-v1", 1000000, 128000),
("global.anthropic.claude-opus-4-7", 1000000, 128000),
(
"global.anthropic.claude-sonnet-4-20250514-v1:0",
1000000,
64000,
),
(
"global.anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
64000,
),
("global.anthropic.claude-sonnet-4-6", 1000000, 64000),
("gmi/anthropic/claude-opus-4", 409600, 32000),
("gmi/anthropic/claude-opus-4.5", 409600, 32000),
("gmi/anthropic/claude-sonnet-4", 409600, 32000),
("gmi/anthropic/claude-sonnet-4.5", 409600, 32000),
("gmi/deepseek-ai/deepseek-v3-0324", 163840, 16384),
("gmi/deepseek-ai/deepseek-v3.2", 163840, 16384),
("gmi/google/gemini-3-flash-preview", 1048576, 65536),
("gmi/google/gemini-3-pro-preview", 1048576, 65536),
("gmi/minimaxai/minimax-m2.1", 196608, 16384),
("gmi/moonshotai/kimi-k2-thinking", 262144, 16384),
("gmi/openai/gpt-4o", 131072, 16384),
("gmi/openai/gpt-4o-mini", 131072, 16384),
("gmi/openai/gpt-5", 409600, 32000),
("gmi/openai/gpt-5.1", 409600, 32000),
("gmi/openai/gpt-5.2", 409600, 32000),
("gmi/qwen/qwen3-vl-235b-a22b-instruct-fp8", 262144, 16384),
("gmi/zai-org/glm-4.7-fp8", 202752, 16384),
("google.gemma-3-12b-it", 128000, 8192),
("google.gemma-3-27b-it", 128000, 8192),
("google.gemma-3-4b-it", 128000, 8192),
("gpt-3.5-turbo", 16385, 4096),
("gpt-3.5-turbo-0125", 16385, 4096),
("gpt-3.5-turbo-1106", 16385, 4096),
("gpt-3.5-turbo-16k", 16385, 4096),
("gpt-3.5-turbo-instruct", 8192, 4096),
("gpt-3.5-turbo-instruct-0914", 8192, 4097),
("gpt-4", 8192, 4096),
("gpt-4-0125-preview", 128000, 4096),
("gpt-4-0314", 8192, 4096),
("gpt-4-0613", 8192, 4096),
("gpt-4-1106-preview", 128000, 4096),
("gpt-4-turbo", 128000, 4096),
("gpt-4-turbo-2024-04-09", 128000, 4096),
("gpt-4-turbo-preview", 128000, 4096),
("gpt-4.1", 1047576, 32768),
("gpt-4.1-2025-04-14", 1047576, 32768),
("gpt-4.1-mini", 1047576, 32768),
("gpt-4.1-mini-2025-04-14", 1047576, 32768),
("gpt-4.1-nano", 1047576, 32768),
("gpt-4.1-nano-2025-04-14", 1047576, 32768),
("gpt-4o", 128000, 16384),
("gpt-4o-2024-05-13", 128000, 4096),
("gpt-4o-2024-08-06", 128000, 16384),
("gpt-4o-2024-11-20", 128000, 16384),
("gpt-4o-audio-preview", 128000, 16384),
("gpt-4o-audio-preview-2024-12-17", 128000, 16384),
("gpt-4o-audio-preview-2025-06-03", 128000, 16384),
("gpt-4o-mini", 128000, 16384),
("gpt-4o-mini-2024-07-18", 128000, 16384),
("gpt-4o-mini-audio-preview", 128000, 16384),
("gpt-4o-mini-audio-preview-2024-12-17", 128000, 16384),
("gpt-4o-mini-realtime-preview", 128000, 4096),
("gpt-4o-mini-realtime-preview-2024-12-17", 128000, 4096),
("gpt-4o-mini-search-preview", 128000, 16384),
("gpt-4o-mini-search-preview-2025-03-11", 128000, 16384),
("gpt-4o-realtime-preview", 128000, 4096),
("gpt-4o-realtime-preview-2024-12-17", 128000, 4096),
("gpt-4o-realtime-preview-2025-06-03", 128000, 4096),
("gpt-4o-search-preview", 128000, 16384),
("gpt-4o-search-preview-2025-03-11", 128000, 16384),
("gpt-5", 272000, 128000),
("gpt-5-2025-08-07", 272000, 128000),
("gpt-5-chat", 128000, 16384),
("gpt-5-chat-latest", 128000, 16384),
("gpt-5-codex", 272000, 128000),
("gpt-5-mini", 272000, 128000),
("gpt-5-mini-2025-08-07", 272000, 128000),
("gpt-5-nano", 272000, 128000),
("gpt-5-nano-2025-08-07", 272000, 128000),
("gpt-5-pro", 128000, 272000),
("gpt-5-pro-2025-10-06", 128000, 272000),
("gpt-5-search-api", 272000, 128000),
("gpt-5-search-api-2025-10-14", 272000, 128000),
("gpt-5.1", 272000, 128000),
("gpt-5.1-2025-11-13", 272000, 128000),
("gpt-5.1-chat-latest", 128000, 16384),
("gpt-5.1-codex", 272000, 128000),
("gpt-5.1-codex-max", 272000, 128000),
("gpt-5.1-codex-mini", 272000, 128000),
("gpt-5.2", 272000, 128000),
("gpt-5.2-2025-12-11", 272000, 128000),
("gpt-5.2-chat-latest", 128000, 16384),
("gpt-5.2-codex", 272000, 128000),
("gpt-5.2-pro", 272000, 128000),
("gpt-5.2-pro-2025-12-11", 272000, 128000),
("gpt-5.3-chat-latest", 128000, 16384),
("gpt-5.3-codex", 272000, 128000),
("gpt-5.4", 1050000, 128000),
("gpt-5.4-2026-03-05", 1050000, 128000),
("gpt-5.4-mini", 272000, 128000),
("gpt-5.4-nano", 272000, 128000),
("gpt-5.4-pro", 1050000, 128000),
("gpt-5.4-pro-2026-03-05", 1050000, 128000),
("gpt-5.5", 1000000, 128000),
("gpt-audio", 128000, 16384),
("gpt-audio-1.5", 128000, 16384),
("gpt-audio-2025-08-28", 128000, 16384),
("gpt-audio-mini", 128000, 16384),
("gpt-audio-mini-2025-10-06", 128000, 16384),
("gpt-audio-mini-2025-12-15", 128000, 16384),
("gpt-realtime", 32000, 4096),
("gpt-realtime-1.5", 32000, 4096),
("gpt-realtime-2025-08-28", 32000, 4096),
("gpt-realtime-mini", 128000, 4096),
("gpt-realtime-mini-2025-10-06", 128000, 4096),
("gpt-realtime-mini-2025-12-15", 128000, 4096),
("gradient_ai/alibaba-qwen3-32b", 2048, 2048),
("gradient_ai/anthropic-claude-3-opus", 1024, 1024),
("gradient_ai/anthropic-claude-3.5-haiku", 1024, 1024),
("gradient_ai/anthropic-claude-3.5-sonnet", 1024, 1024),
("gradient_ai/anthropic-claude-3.7-sonnet", 1024, 1024),
("gradient_ai/deepseek-r1-distill-llama-70b", 8000, 8000),
("gradient_ai/llama3-8b-instruct", 512, 512),
("gradient_ai/llama3.3-70b-instruct", 2048, 2048),
("gradient_ai/mistral-nemo-instruct-2407", 512, 512),
("gradient_ai/openai-gpt-4o", 16384, 16384),
("gradient_ai/openai-gpt-4o-mini", 16384, 16384),
("gradient_ai/openai-o3", 100000, 100000),
("gradient_ai/openai-o3-mini", 100000, 100000),
("groq/gemma-7b-it", 8192, 8192),
("groq/llama-3.1-8b-instant", 128000, 8192),
("groq/llama-3.3-70b-versatile", 128000, 32768),
(
"groq/meta-llama/llama-4-maverick-17b-128e-instruct",
131072,
8192,
),
(
"groq/meta-llama/llama-4-scout-17b-16e-instruct",
131072,
8192,
),
("groq/meta-llama/llama-guard-4-12b", 8192, 8192),
("groq/moonshotai/kimi-k2-instruct-0905", 262144, 16384),
("groq/openai/gpt-oss-120b", 131072, 32766),
("groq/openai/gpt-oss-20b", 131072, 32768),
("groq/openai/gpt-oss-safeguard-20b", 131072, 65536),
("groq/qwen/qwen3-32b", 131000, 131000),
("heroku/claude-3-5-haiku", 4096, 4096),
("heroku/claude-3-5-sonnet-latest", 8192, 8192),
("heroku/claude-3-7-sonnet", 8192, 8192),
("heroku/claude-4-sonnet", 8192, 8192),
("hyperbolic/deepseek-ai/deepseek-r1", 32768, 32768),
("hyperbolic/deepseek-ai/deepseek-r1-0528", 131072, 131072),
("hyperbolic/deepseek-ai/deepseek-v3", 32768, 32768),
("hyperbolic/deepseek-ai/deepseek-v3-0324", 32768, 32768),
("hyperbolic/meta-llama/llama-3.2-3b-instruct", 32768, 32768),
(
"hyperbolic/meta-llama/llama-3.3-70b-instruct",
131072,
131072,
),
(
"hyperbolic/meta-llama/meta-llama-3-70b-instruct",
131072,
131072,
),
(
"hyperbolic/meta-llama/meta-llama-3.1-405b-instruct",
32768,
32768,
),
(
"hyperbolic/meta-llama/meta-llama-3.1-70b-instruct",
32768,
32768,
),
(
"hyperbolic/meta-llama/meta-llama-3.1-8b-instruct",
32768,
32768,
),
("hyperbolic/moonshotai/kimi-k2-instruct", 131072, 131072),
(
"hyperbolic/nousresearch/hermes-3-llama-3.1-70b",
32768,
32768,
),
("hyperbolic/qwen/qwen2.5-72b-instruct", 131072, 131072),
("hyperbolic/qwen/qwen2.5-coder-32b-instruct", 32768, 32768),
("hyperbolic/qwen/qwen3-235b-a22b", 131072, 131072),
("hyperbolic/qwen/qwq-32b", 131072, 131072),
("j2-light", 8192, 8192),
("j2-mid", 8192, 8192),
("j2-ultra", 8192, 8192),
("jamba-1.5", 256000, 256000),
("jamba-1.5-large", 256000, 256000),
("jamba-1.5-large@001", 256000, 256000),
("jamba-1.5-mini", 256000, 256000),
("jamba-1.5-mini@001", 256000, 256000),
("jamba-large-1.6", 256000, 256000),
("jamba-large-1.7", 256000, 256000),
("jamba-mini-1.6", 256000, 256000),
("jamba-mini-1.7", 256000, 256000),
("jp.anthropic.claude-haiku-4-5-20251001-v1:0", 200000, 64000),
(
"jp.anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
64000,
),
("kimi-k2-thinking-251104", 229376, 32768),
("lambda_ai/deepseek-llama3.3-70b", 131072, 131072),
("lambda_ai/deepseek-r1-0528", 131072, 131072),
("lambda_ai/deepseek-r1-671b", 131072, 131072),
("lambda_ai/deepseek-v3-0324", 131072, 131072),
("lambda_ai/hermes3-405b", 131072, 131072),
("lambda_ai/hermes3-70b", 131072, 131072),
("lambda_ai/hermes3-8b", 131072, 131072),
("lambda_ai/lfm-40b", 131072, 131072),
("lambda_ai/lfm-7b", 131072, 131072),
(
"lambda_ai/llama-4-maverick-17b-128e-instruct-fp8",
131072,
8192,
),
("lambda_ai/llama-4-scout-17b-16e-instruct", 16384, 8192),
("lambda_ai/llama3.1-405b-instruct-fp8", 131072, 131072),
("lambda_ai/llama3.1-70b-instruct-fp8", 131072, 131072),
("lambda_ai/llama3.1-8b-instruct", 131072, 131072),
(
"lambda_ai/llama3.1-nemotron-70b-instruct-fp8",
131072,
131072,
),
("lambda_ai/llama3.2-11b-vision-instruct", 131072, 131072),
("lambda_ai/llama3.2-3b-instruct", 131072, 131072),
("lambda_ai/llama3.3-70b-instruct-fp8", 131072, 131072),
("lambda_ai/qwen25-coder-32b-instruct", 131072, 131072),
("lambda_ai/qwen3-32b-fp8", 131072, 131072),
("lemonade/gemma-3-4b-it-gguf", 128000, 8192),
("lemonade/gpt-oss-120b-mxfp-gguf", 131072, 32768),
("lemonade/gpt-oss-20b-mxfp4-gguf", 131072, 32768),
("lemonade/qwen3-4b-instruct-2507-gguf", 262144, 32768),
("lemonade/qwen3-coder-30b-a3b-instruct-gguf", 262144, 32768),
("llamagate/codellama-7b", 16384, 4096),
("llamagate/deepseek-coder-6.7b", 16384, 4096),
("llamagate/deepseek-r1-7b-qwen", 131072, 16384),
("llamagate/deepseek-r1-8b", 65536, 16384),
("llamagate/dolphin3-8b", 128000, 8192),
("llamagate/gemma3-4b", 128000, 8192),
("llamagate/llama-3.1-8b", 131072, 8192),
("llamagate/llama-3.2-3b", 131072, 8192),
("llamagate/llava-7b", 4096, 2048),
("llamagate/mistral-7b-v0.3", 32768, 8192),
("llamagate/openthinker-7b", 32768, 8192),
("llamagate/qwen2.5-coder-7b", 32768, 8192),
("llamagate/qwen3-8b", 32768, 8192),
("llamagate/qwen3-vl-8b", 32768, 8192),
("luminous-base", 2048, 2048),
("luminous-base-control", 2048, 2048),
("luminous-extended", 2048, 2048),
("luminous-extended-control", 2048, 2048),
("luminous-supreme", 2048, 2048),
("luminous-supreme-control", 2048, 2048),
("medlm-large", 8192, 1024),
("medlm-medium", 32768, 8192),
("meta.llama2-13b-chat-v1", 4096, 4096),
("meta.llama2-70b-chat-v1", 4096, 4096),
("meta.llama3-1-405b-instruct-v1:0", 128000, 4096),
("meta.llama3-1-70b-instruct-v1:0", 128000, 2048),
("meta.llama3-1-8b-instruct-v1:0", 128000, 2048),
("meta.llama3-2-11b-instruct-v1:0", 128000, 4096),
("meta.llama3-2-1b-instruct-v1:0", 128000, 4096),
("meta.llama3-2-3b-instruct-v1:0", 128000, 4096),
("meta.llama3-2-90b-instruct-v1:0", 128000, 4096),
("meta.llama3-3-70b-instruct-v1:0", 128000, 4096),
("meta.llama3-70b-instruct-v1:0", 8192, 8192),
("meta.llama3-8b-instruct-v1:0", 8192, 8192),
("meta.llama4-maverick-17b-instruct-v1:0", 128000, 4096),
("meta.llama4-scout-17b-instruct-v1:0", 128000, 4096),
("meta_llama/llama-3.3-70b-instruct", 128000, 4028),
("meta_llama/llama-3.3-8b-instruct", 128000, 4028),
(
"meta_llama/llama-4-maverick-17b-128e-instruct-fp8",
1000000,
4028,
),
(
"meta_llama/llama-4-scout-17b-16e-instruct-fp8",
10000000,
4028,
),
("minimax.minimax-m2", 128000, 8192),
("minimax.minimax-m2.1", 196000, 8192),
("minimax.minimax-m2.5", 1000000, 8192),
("minimax/minimax-m2", 200000, 8192),
("minimax/minimax-m2.1", 1000000, 8192),
("minimax/minimax-m2.1-lightning", 1000000, 8192),
("minimax/minimax-m2.5", 1000000, 8192),
("minimax/minimax-m2.5-lightning", 1000000, 8192),
("mistral.devstral-2-123b", 256000, 8192),
("mistral.magistral-small-2509", 128000, 8192),
("mistral.ministral-3-14b-instruct", 128000, 8192),
("mistral.ministral-3-3b-instruct", 128000, 8192),
("mistral.ministral-3-8b-instruct", 128000, 8192),
("mistral.mistral-7b-instruct-v0:2", 32000, 8191),
("mistral.mistral-large-2402-v1:0", 32000, 8191),
("mistral.mistral-large-2407-v1:0", 128000, 8191),
("mistral.mistral-large-3-675b-instruct", 128000, 8192),
("mistral.mistral-small-2402-v1:0", 32000, 8191),
("mistral.mixtral-8x7b-instruct-v0:1", 32000, 8191),
("mistral.voxtral-mini-3b-2507", 128000, 8192),
("mistral.voxtral-small-24b-2507", 128000, 8192),
("mistral/codestral-2405", 32000, 8191),
("mistral/codestral-2508", 256000, 256000),
("mistral/codestral-latest", 32000, 8191),
("mistral/codestral-mamba-latest", 256000, 256000),
("mistral/devstral-2512", 256000, 256000),
("mistral/devstral-latest", 256000, 256000),
("mistral/devstral-medium-2507", 128000, 128000),
("mistral/devstral-medium-latest", 256000, 256000),
("mistral/devstral-small-2505", 128000, 128000),
("mistral/devstral-small-2507", 128000, 128000),
("mistral/devstral-small-latest", 256000, 256000),
("mistral/labs-devstral-small-2512", 256000, 256000),
("mistral/magistral-medium-1-2-2509", 40000, 40000),
("mistral/magistral-medium-2506", 40000, 40000),
("mistral/magistral-medium-2509", 40000, 40000),
("mistral/magistral-medium-latest", 40000, 40000),
("mistral/magistral-small-1-2-2509", 40000, 40000),
("mistral/magistral-small-2506", 40000, 40000),
("mistral/magistral-small-latest", 40000, 40000),
("mistral/ministral-3-14b-2512", 262144, 262144),
("mistral/ministral-3-3b-2512", 131072, 131072),
("mistral/ministral-3-8b-2512", 262144, 262144),
("mistral/mistral-large-2402", 32000, 8191),
("mistral/mistral-large-2407", 128000, 128000),
("mistral/mistral-large-2411", 128000, 128000),
("mistral/mistral-large-2512", 262144, 262144),
("mistral/mistral-large-3", 262144, 262144),
("mistral/mistral-large-latest", 262144, 262144),
("mistral/mistral-medium", 32000, 8191),
("mistral/mistral-medium-2312", 32000, 8191),
("mistral/mistral-medium-2505", 131072, 8191),
("mistral/mistral-medium-3-1-2508", 131072, 131072),
("mistral/mistral-medium-latest", 131072, 131072),
("mistral/mistral-small", 32000, 8191),
("mistral/mistral-small-3-2-2506", 131072, 131072),
("mistral/mistral-small-latest", 131072, 131072),
("mistral/mistral-tiny", 32000, 8191),
("mistral/open-codestral-mamba", 256000, 256000),
("mistral/open-mistral-7b", 32000, 8191),
("mistral/open-mistral-nemo", 128000, 128000),
("mistral/open-mistral-nemo-2407", 128000, 128000),
("mistral/open-mixtral-8x22b", 65336, 8191),
("mistral/open-mixtral-8x7b", 32000, 8191),
("mistral/pixtral-12b-2409", 128000, 128000),
("mistral/pixtral-large-2411", 128000, 128000),
("mistral/pixtral-large-latest", 128000, 128000),
("moonshot.kimi-k2-thinking", 128000, 8192),
("moonshot/kimi-k2-0711-preview", 131072, 131072),
("moonshot/kimi-k2-0905-preview", 262144, 262144),
("moonshot/kimi-k2-thinking", 262144, 262144),
("moonshot/kimi-k2-thinking-turbo", 262144, 262144),
("moonshot/kimi-k2-turbo-preview", 262144, 262144),
("moonshot/kimi-k2.5", 262144, 262144),
("moonshot/kimi-k2.6", 262144, 262144),
("moonshot/kimi-latest", 131072, 131072),
("moonshot/kimi-latest-128k", 131072, 131072),
("moonshot/kimi-latest-32k", 32768, 32768),
("moonshot/kimi-latest-8k", 8192, 8192),
("moonshot/kimi-thinking-preview", 131072, 131072),
("moonshot/moonshot-v1-128k", 131072, 131072),
("moonshot/moonshot-v1-128k-0430", 131072, 131072),
("moonshot/moonshot-v1-128k-vision-preview", 131072, 131072),
("moonshot/moonshot-v1-32k", 32768, 32768),
("moonshot/moonshot-v1-32k-0430", 32768, 32768),
("moonshot/moonshot-v1-32k-vision-preview", 32768, 32768),
("moonshot/moonshot-v1-8k", 8192, 8192),
("moonshot/moonshot-v1-8k-0430", 8192, 8192),
("moonshot/moonshot-v1-8k-vision-preview", 8192, 8192),
("moonshot/moonshot-v1-auto", 131072, 131072),
("moonshotai.kimi-k2.5", 262144, 262144),
("morph/morph-v3-fast", 16000, 16000),
("morph/morph-v3-large", 16000, 16000),
("nebius/deepseek-ai/deepseek-r1", 128000, 128000),
("nebius/deepseek-ai/deepseek-r1-0528", 164000, 164000),
(
"nebius/deepseek-ai/deepseek-r1-distill-llama-70b",
128000,
128000,
),
("nebius/deepseek-ai/deepseek-v3", 128000, 128000),
("nebius/deepseek-ai/deepseek-v3-0324", 128000, 128000),
("nebius/google/gemma-3-27b-it", 128000, 128000),
("nebius/meta-llama/llama-3.3-70b-instruct", 128000, 128000),
("nebius/meta-llama/llama-guard-3-8b", 128000, 128000),
(
"nebius/meta-llama/meta-llama-3.1-405b-instruct",
128000,
128000,
),
(
"nebius/meta-llama/meta-llama-3.1-70b-instruct",
128000,
128000,
),
(
"nebius/meta-llama/meta-llama-3.1-8b-instruct",
128000,
128000,
),
(
"nebius/mistralai/mistral-nemo-instruct-2407",
128000,
128000,
),
(
"nebius/nousresearch/hermes-3-llama-3.1-405b",
128000,
128000,
),
(
"nebius/nvidia/llama-3.1-nemotron-ultra-253b-v1",
128000,
128000,
),
(
"nebius/nvidia/llama-3.3-nemotron-super-49b-v1",
131072,
131072,
),
("nebius/qwen/qwen2-vl-72b-instruct", 131072, 131072),
("nebius/qwen/qwen2-vl-7b-instruct", 131072, 131072),
("nebius/qwen/qwen2.5-32b-instruct", 128000, 128000),
("nebius/qwen/qwen2.5-72b-instruct", 128000, 128000),
("nebius/qwen/qwen2.5-coder-7b", 32768, 32768),
("nebius/qwen/qwen2.5-vl-72b-instruct", 131072, 131072),
("nebius/qwen/qwen3-14b", 32768, 32768),
("nebius/qwen/qwen3-235b-a22b", 262144, 262144),
("nebius/qwen/qwen3-30b-a3b", 32768, 32768),
("nebius/qwen/qwen3-32b", 32768, 32768),
("nebius/qwen/qwen3-4b", 32768, 32768),
("nebius/qwen/qwq-32b", 32768, 32768),
("novita/baichuan/baichuan-m2-32b", 131072, 131072),
("novita/baidu/ernie-4.5-21b-a3b", 120000, 8000),
("novita/baidu/ernie-4.5-21b-a3b-thinking", 131072, 65536),
("novita/baidu/ernie-4.5-300b-a47b-paddle", 123000, 12000),
("novita/baidu/ernie-4.5-vl-28b-a3b", 30000, 8000),
("novita/baidu/ernie-4.5-vl-28b-a3b-thinking", 131072, 65536),
("novita/baidu/ernie-4.5-vl-424b-a47b", 123000, 16000),
("novita/deepseek/deepseek-ocr", 8192, 8192),
("novita/deepseek/deepseek-prover-v2-671b", 160000, 160000),
("novita/deepseek/deepseek-r1-0528", 163840, 32768),
("novita/deepseek/deepseek-r1-0528-qwen3-8b", 128000, 32000),
("novita/deepseek/deepseek-r1-distill-llama-70b", 8192, 8192),
("novita/deepseek/deepseek-r1-distill-qwen-14b", 32768, 16384),
("novita/deepseek/deepseek-r1-distill-qwen-32b", 64000, 32000),
("novita/deepseek/deepseek-r1-turbo", 64000, 16000),
("novita/deepseek/deepseek-v3-0324", 163840, 163840),
("novita/deepseek/deepseek-v3-turbo", 64000, 16000),
("novita/deepseek/deepseek-v3.1", 131072, 32768),
("novita/deepseek/deepseek-v3.1-terminus", 131072, 32768),
("novita/deepseek/deepseek-v3.2", 163840, 65536),
("novita/deepseek/deepseek-v3.2-exp", 163840, 65536),
("novita/google/gemma-3-12b-it", 131072, 8192),
("novita/google/gemma-3-27b-it", 98304, 16384),
("novita/gryphe/mythomax-l2-13b", 4096, 3200),
("novita/kwaipilot/kat-coder-pro", 256000, 128000),
("novita/meta-llama/llama-3-70b-instruct", 8192, 8000),
("novita/meta-llama/llama-3-8b-instruct", 8192, 8192),
("novita/meta-llama/llama-3.1-8b-instruct", 16384, 16384),
("novita/meta-llama/llama-3.2-3b-instruct", 32768, 32000),
("novita/meta-llama/llama-3.3-70b-instruct", 131072, 120000),
(
"novita/meta-llama/llama-4-maverick-17b-128e-instruct-fp8",
1048576,
8192,
),
(
"novita/meta-llama/llama-4-scout-17b-16e-instruct",
131072,
131072,
),
("novita/microsoft/wizardlm-2-8x22b", 65535, 8000),
("novita/minimax/minimax-m2", 204800, 131072),
("novita/minimax/minimax-m2.1", 204800, 131072),
("novita/minimaxai/minimax-m1-80k", 1000000, 40000),
("novita/mistralai/mistral-nemo", 60288, 16000),
("novita/moonshotai/kimi-k2-0905", 262144, 262144),
("novita/moonshotai/kimi-k2-instruct", 131072, 131072),
("novita/moonshotai/kimi-k2-thinking", 262144, 262144),
("novita/nousresearch/hermes-2-pro-llama-3-8b", 8192, 8192),
("novita/openai/gpt-oss-120b", 131072, 32768),
("novita/openai/gpt-oss-20b", 131072, 32768),
("novita/paddlepaddle/paddleocr-vl", 16384, 16384),
("novita/qwen/qwen-2.5-72b-instruct", 32000, 8192),
("novita/qwen/qwen-mt-plus", 16384, 8192),
("novita/qwen/qwen2.5-7b-instruct", 32000, 32000),
("novita/qwen/qwen2.5-vl-72b-instruct", 32768, 32768),
("novita/qwen/qwen3-235b-a22b-fp8", 40960, 20000),
("novita/qwen/qwen3-235b-a22b-instruct-2507", 131072, 16384),
("novita/qwen/qwen3-235b-a22b-thinking-2507", 131072, 32768),
("novita/qwen/qwen3-30b-a3b-fp8", 40960, 20000),
("novita/qwen/qwen3-32b-fp8", 40960, 20000),
("novita/qwen/qwen3-4b-fp8", 128000, 20000),
("novita/qwen/qwen3-8b-fp8", 128000, 20000),
("novita/qwen/qwen3-coder-30b-a3b-instruct", 160000, 32768),
("novita/qwen/qwen3-coder-480b-a35b-instruct", 262144, 65536),
("novita/qwen/qwen3-max", 262144, 65536),
("novita/qwen/qwen3-next-80b-a3b-instruct", 131072, 32768),
("novita/qwen/qwen3-next-80b-a3b-thinking", 131072, 32768),
("novita/qwen/qwen3-omni-30b-a3b-instruct", 65536, 16384),
("novita/qwen/qwen3-omni-30b-a3b-thinking", 65536, 16384),
("novita/qwen/qwen3-vl-235b-a22b-instruct", 131072, 32768),
("novita/qwen/qwen3-vl-235b-a22b-thinking", 131072, 32768),
("novita/qwen/qwen3-vl-30b-a3b-instruct", 131072, 32768),
("novita/qwen/qwen3-vl-30b-a3b-thinking", 131072, 32768),
("novita/qwen/qwen3-vl-8b-instruct", 131072, 32768),
("novita/sao10k/l3-70b-euryale-v2.1", 8192, 8192),
("novita/sao10k/l3-8b-lunaris", 8192, 8192),
("novita/sao10k/l3-8b-stheno-v3.2", 8192, 32000),
("novita/sao10k/l31-70b-euryale-v2.2", 8192, 8192),
("novita/skywork/r1v4-lite", 262144, 65536),
("novita/xiaomimimo/mimo-v2-flash", 262144, 32000),
("novita/zai-org/autoglm-phone-9b-multilingual", 65536, 65536),
("novita/zai-org/glm-4.5", 131072, 98304),
("novita/zai-org/glm-4.5-air", 131072, 98304),
("novita/zai-org/glm-4.5v", 65536, 16384),
("novita/zai-org/glm-4.6", 204800, 131072),
("novita/zai-org/glm-4.6v", 131072, 32768),
("novita/zai-org/glm-4.7", 204800, 131072),
("nvidia.nemotron-nano-12b-v2", 128000, 8192),
("nvidia.nemotron-nano-3-30b", 262144, 8192),
("nvidia.nemotron-nano-9b-v2", 128000, 8192),
("nvidia.nemotron-super-3-120b", 256000, 32768),
("o1", 200000, 100000),
("o1-2024-12-17", 200000, 100000),
("o1-pro", 200000, 100000),
("o1-pro-2025-03-19", 200000, 100000),
("o3", 200000, 100000),
("o3-2025-04-16", 200000, 100000),
("o3-deep-research", 200000, 100000),
("o3-deep-research-2025-06-26", 200000, 100000),
("o3-mini", 200000, 100000),
("o3-mini-2025-01-31", 200000, 100000),
("o3-pro", 200000, 100000),
("o3-pro-2025-06-10", 200000, 100000),
("o4-mini", 200000, 100000),
("o4-mini-2025-04-16", 200000, 100000),
("o4-mini-deep-research", 200000, 100000),
("o4-mini-deep-research-2025-06-26", 200000, 100000),
("oci/cohere.command-a-03-2025", 256000, 4000),
("oci/cohere.command-a-reasoning-08-2025", 256000, 4000),
("oci/cohere.command-a-translate-08-2025", 256000, 4000),
("oci/cohere.command-a-vision-07-2025", 128000, 4000),
("oci/cohere.command-latest", 128000, 4000),
("oci/cohere.command-plus-latest", 128000, 4000),
("oci/cohere.command-r-08-2024", 128000, 4000),
("oci/cohere.command-r-plus-08-2024", 128000, 4000),
("oci/google.gemini-2.5-flash", 1048576, 65536),
("oci/google.gemini-2.5-flash-lite", 1048576, 65536),
("oci/google.gemini-2.5-pro", 1048576, 65536),
("oci/meta.llama-3.1-405b-instruct", 128000, 4000),
("oci/meta.llama-3.1-70b-instruct", 128000, 4000),
("oci/meta.llama-3.2-11b-vision-instruct", 128000, 4000),
("oci/meta.llama-3.2-90b-vision-instruct", 128000, 4000),
("oci/meta.llama-3.3-70b-instruct", 128000, 4000),
("oci/meta.llama-3.3-70b-instruct-fp8-dynamic", 128000, 4000),
(
"oci/meta.llama-4-maverick-17b-128e-instruct-fp8",
512000,
4000,
),
("oci/meta.llama-4-scout-17b-16e-instruct", 192000, 4000),
("oci/xai.grok-3", 131072, 131072),
("oci/xai.grok-3-fast", 131072, 131072),
("oci/xai.grok-3-mini", 131072, 131072),
("oci/xai.grok-3-mini-fast", 131072, 131072),
("oci/xai.grok-4", 128000, 128000),
("oci/xai.grok-4-fast", 131072, 131072),
("oci/xai.grok-4.1-fast", 131072, 131072),
("oci/xai.grok-4.20", 131072, 131072),
("oci/xai.grok-4.20-multi-agent", 131072, 131072),
("oci/xai.grok-code-fast-1", 131072, 131072),
("ollama/codegeex4", 32768, 8192),
("ollama/codegemma", 8192, 8192),
("ollama/codellama", 4096, 4096),
("ollama/deepseek-coder-v2-base", 8192, 8192),
("ollama/deepseek-coder-v2-instruct", 32768, 8192),
("ollama/deepseek-coder-v2-lite-base", 8192, 8192),
("ollama/deepseek-coder-v2-lite-instruct", 32768, 8192),
("ollama/deepseek-v3.1:671b-cloud", 163840, 163840),
("ollama/gpt-oss:120b-cloud", 131072, 131072),
("ollama/gpt-oss:20b-cloud", 131072, 131072),
("ollama/internlm2_5-20b-chat", 32768, 8192),
("ollama/llama2", 4096, 4096),
("ollama/llama2-uncensored", 4096, 4096),
("ollama/llama2:13b", 4096, 4096),
("ollama/llama2:70b", 4096, 4096),
("ollama/llama2:7b", 4096, 4096),
("ollama/llama3", 8192, 8192),
("ollama/llama3.1", 8192, 8192),
("ollama/llama3:70b", 8192, 8192),
("ollama/llama3:8b", 8192, 8192),
("ollama/mistral", 8192, 8192),
("ollama/mistral-7b-instruct-v0.1", 8192, 8192),
("ollama/mistral-7b-instruct-v0.2", 32768, 32768),
("ollama/mistral-large-instruct-2407", 65536, 8192),
("ollama/mixtral-8x22b-instruct-v0.1", 65536, 65536),
("ollama/mixtral-8x7b-instruct-v0.1", 32768, 32768),
("ollama/orca-mini", 4096, 4096),
("ollama/qwen3-coder:480b-cloud", 262144, 262144),
("ollama/vicuna", 2048, 2048),
("openai.gpt-oss-120b-1:0", 128000, 128000),
("openai.gpt-oss-20b-1:0", 128000, 128000),
("openai.gpt-oss-safeguard-120b", 128000, 8192),
("openai.gpt-oss-safeguard-20b", 128000, 8192),
("openrouter/anthropic/claude-3-haiku", 200000, 200000),
("openrouter/anthropic/claude-3.5-sonnet", 200000, 8192),
("openrouter/anthropic/claude-3.7-sonnet", 200000, 128000),
("openrouter/anthropic/claude-haiku-4.5", 200000, 200000),
("openrouter/anthropic/claude-opus-4", 200000, 32000),
("openrouter/anthropic/claude-opus-4.1", 200000, 32000),
("openrouter/anthropic/claude-opus-4.5", 200000, 32000),
("openrouter/anthropic/claude-opus-4.6", 1000000, 128000),
("openrouter/anthropic/claude-opus-4.7", 1000000, 128000),
("openrouter/anthropic/claude-sonnet-4", 1000000, 64000),
("openrouter/anthropic/claude-sonnet-4.5", 1000000, 1000000),
("openrouter/anthropic/claude-sonnet-4.6", 1000000, 128000),
("openrouter/bytedance/ui-tars-1.5-7b", 131072, 2048),
("openrouter/deepseek/deepseek-chat", 65536, 8192),
("openrouter/deepseek/deepseek-chat-v3-0324", 65536, 8192),
("openrouter/deepseek/deepseek-chat-v3.1", 163840, 163840),
("openrouter/deepseek/deepseek-r1", 65336, 8192),
("openrouter/deepseek/deepseek-r1-0528", 65336, 8192),
("openrouter/deepseek/deepseek-v3.2", 163840, 163840),
("openrouter/deepseek/deepseek-v3.2-exp", 163840, 163840),
("openrouter/google/gemini-2.0-flash-001", 1048576, 8192),
("openrouter/google/gemini-2.5-flash", 1048576, 8192),
("openrouter/google/gemini-2.5-pro", 1048576, 8192),
("openrouter/google/gemini-3-flash-preview", 1048576, 65535),
("openrouter/google/gemini-3-pro-preview", 1048576, 65535),
(
"openrouter/google/gemini-3.1-flash-lite-preview",
1048576,
65536,
),
("openrouter/google/gemini-3.1-pro-preview", 1048576, 65536),
("openrouter/gryphe/mythomax-l2-13b", 8192, 8192),
("openrouter/mancer/weaver", 8000, 8000),
("openrouter/meta-llama/llama-3-70b-instruct", 8192, 8192),
("openrouter/minimax/minimax-m2", 204800, 204800),
("openrouter/minimax/minimax-m2.1", 204000, 64000),
("openrouter/minimax/minimax-m2.5", 196608, 65536),
("openrouter/mistralai/devstral-2512", 262144, 65536),
("openrouter/mistralai/ministral-14b-2512", 262144, 262144),
("openrouter/mistralai/ministral-3b-2512", 131072, 131072),
("openrouter/mistralai/ministral-8b-2512", 262144, 262144),
("openrouter/mistralai/mistral-7b-instruct", 8192, 8192),
("openrouter/mistralai/mistral-large", 32000, 32000),
("openrouter/mistralai/mistral-large-2512", 262144, 262144),
(
"openrouter/mistralai/mistral-small-3.1-24b-instruct",
32000,
32000,
),
(
"openrouter/mistralai/mistral-small-3.2-24b-instruct",
32000,
32000,
),
("openrouter/mistralai/mixtral-8x22b-instruct", 65536, 65536),
("openrouter/moonshotai/kimi-k2.5", 262144, 262144),
("openrouter/openai/gpt-3.5-turbo", 4095, 4095),
("openrouter/openai/gpt-3.5-turbo-16k", 16383, 16383),
("openrouter/openai/gpt-4", 8192, 8192),
("openrouter/openai/gpt-4.1", 1047576, 32768),
("openrouter/openai/gpt-4.1-mini", 1047576, 32768),
("openrouter/openai/gpt-4.1-nano", 1047576, 32768),
("openrouter/openai/gpt-4o", 128000, 4096),
("openrouter/openai/gpt-4o-2024-05-13", 128000, 4096),
("openrouter/openai/gpt-5", 272000, 128000),
("openrouter/openai/gpt-5-chat", 128000, 16384),
("openrouter/openai/gpt-5-codex", 272000, 128000),
("openrouter/openai/gpt-5-mini", 272000, 128000),
("openrouter/openai/gpt-5-nano", 272000, 128000),
("openrouter/openai/gpt-5.1-codex-max", 400000, 128000),
("openrouter/openai/gpt-5.2", 272000, 128000),
("openrouter/openai/gpt-5.2-chat", 128000, 16384),
("openrouter/openai/gpt-5.2-codex", 272000, 128000),
("openrouter/openai/gpt-5.2-pro", 272000, 128000),
("openrouter/openai/gpt-oss-120b", 131072, 32768),
("openrouter/openai/gpt-oss-20b", 131072, 32768),
("openrouter/openai/o1", 200000, 100000),
("openrouter/openai/o3-mini", 128000, 65536),
("openrouter/openai/o3-mini-high", 128000, 65536),
("openrouter/openrouter/auto", 2000000, 2000000),
("openrouter/openrouter/bodybuilder", 128000, 128000),
("openrouter/openrouter/free", 200000, 200000),
("openrouter/qwen/qwen-2.5-coder-32b-instruct", 33792, 33792),
("openrouter/qwen/qwen-vl-plus", 8192, 2048),
("openrouter/qwen/qwen3-235b-a22b-2507", 262144, 262144),
(
"openrouter/qwen/qwen3-235b-a22b-thinking-2507",
262144,
262144,
),
("openrouter/qwen/qwen3-coder", 262100, 262100),
("openrouter/qwen/qwen3-coder-plus", 997952, 65536),
("openrouter/qwen/qwen3.5-122b-a10b", 262144, 65536),
("openrouter/qwen/qwen3.5-27b", 262144, 65536),
("openrouter/qwen/qwen3.5-35b-a3b", 262144, 65536),
("openrouter/qwen/qwen3.5-397b-a17b", 262144, 65536),
("openrouter/qwen/qwen3.5-flash-02-23", 1000000, 65536),
("openrouter/qwen/qwen3.5-plus-02-15", 1000000, 65536),
("openrouter/switchpoint/router", 131072, 131072),
("openrouter/undi95/remm-slerp-l2-13b", 6144, 6144),
("openrouter/x-ai/grok-4", 256000, 256000),
("openrouter/xiaomi/mimo-v2-flash", 262144, 16384),
("openrouter/z-ai/glm-4.6", 202800, 131000),
("openrouter/z-ai/glm-4.6:exacto", 202800, 131000),
("openrouter/z-ai/glm-4.7", 202752, 64000),
("openrouter/z-ai/glm-4.7-flash", 200000, 32000),
("openrouter/z-ai/glm-5", 202752, 128000),
("ovhcloud/deepseek-r1-distill-llama-70b", 131000, 131000),
("ovhcloud/gpt-oss-120b", 131000, 131000),
("ovhcloud/gpt-oss-20b", 131000, 131000),
("ovhcloud/llama-3.1-8b-instruct", 131000, 131000),
("ovhcloud/llava-v1.6-mistral-7b-hf", 32000, 32000),
("ovhcloud/mamba-codestral-7b-v0.1", 256000, 256000),
("ovhcloud/meta-llama-3_1-70b-instruct", 131000, 131000),
("ovhcloud/meta-llama-3_3-70b-instruct", 131000, 131000),
("ovhcloud/mistral-7b-instruct-v0.3", 127000, 127000),
("ovhcloud/mistral-nemo-instruct-2407", 118000, 118000),
(
"ovhcloud/mistral-small-3.2-24b-instruct-2506",
128000,
128000,
),
("ovhcloud/mixtral-8x7b-instruct-v0.1", 32000, 32000),
("ovhcloud/qwen2.5-coder-32b-instruct", 32000, 32000),
("ovhcloud/qwen2.5-vl-72b-instruct", 32000, 32000),
("ovhcloud/qwen3-32b", 32000, 32000),
("palm/chat-bison", 8192, 4096),
("palm/chat-bison-001", 8192, 4096),
("palm/text-bison", 8192, 1024),
("palm/text-bison-001", 8192, 1024),
("palm/text-bison-safety-off", 8192, 1024),
("palm/text-bison-safety-recitation-off", 8192, 1024),
("perplexity/codellama-34b-instruct", 16384, 16384),
("perplexity/codellama-70b-instruct", 16384, 16384),
("perplexity/llama-2-70b-chat", 4096, 4096),
("perplexity/llama-3.1-70b-instruct", 131072, 131072),
("perplexity/llama-3.1-8b-instruct", 131072, 131072),
("perplexity/mistral-7b-instruct", 4096, 4096),
("perplexity/mixtral-8x7b-instruct", 4096, 4096),
("perplexity/pplx-70b-chat", 4096, 4096),
("perplexity/pplx-70b-online", 4096, 4096),
("perplexity/pplx-7b-chat", 8192, 8192),
("perplexity/pplx-7b-online", 4096, 4096),
("perplexity/sonar", 128000, 128000),
("perplexity/sonar-deep-research", 128000, 128000),
("perplexity/sonar-medium-chat", 16384, 16384),
("perplexity/sonar-medium-online", 12000, 12000),
("perplexity/sonar-pro", 200000, 8000),
("perplexity/sonar-reasoning", 128000, 128000),
("perplexity/sonar-reasoning-pro", 128000, 128000),
("perplexity/sonar-small-chat", 16384, 16384),
("perplexity/sonar-small-online", 12000, 12000),
("publicai/aisingapore/gemma-sea-lion-v4-27b-it", 8192, 4096),
("publicai/aisingapore/qwen-sea-lion-v4-32b-it", 32768, 4096),
("publicai/allenai/olmo-3-32b-think", 32768, 4096),
("publicai/allenai/olmo-3-7b-instruct", 32768, 4096),
("publicai/allenai/olmo-3-7b-think", 32768, 4096),
("publicai/bsc-lt/alia-40b-instruct_q8_0", 8192, 4096),
(
"publicai/bsc-lt/salamandra-7b-instruct-tools-16k",
16384,
4096,
),
("publicai/swiss-ai/apertus-70b-instruct", 8192, 4096),
("publicai/swiss-ai/apertus-8b-instruct", 8192, 4096),
("qwen.qwen3-235b-a22b-2507-v1:0", 262144, 131072),
("qwen.qwen3-32b-v1:0", 131072, 16384),
("qwen.qwen3-coder-30b-a3b-v1:0", 262144, 131072),
("qwen.qwen3-coder-480b-a35b-v1:0", 262000, 65536),
("qwen.qwen3-coder-next", 262144, 8192),
("qwen.qwen3-next-80b-a3b", 128000, 8192),
("qwen.qwen3-vl-235b-a22b", 128000, 8192),
("replicate/deepseek-ai/deepseek-r1", 65536, 8192),
("replicate/deepseek-ai/deepseek-v3", 65536, 8192),
("replicate/deepseek-ai/deepseek-v3.1", 163840, 163840),
("replicate/meta/llama-2-13b", 4096, 4096),
("replicate/meta/llama-2-13b-chat", 4096, 4096),
("replicate/meta/llama-2-70b", 4096, 4096),
("replicate/meta/llama-2-70b-chat", 4096, 4096),
("replicate/meta/llama-2-7b", 4096, 4096),
("replicate/meta/llama-2-7b-chat", 4096, 4096),
("replicate/meta/llama-3-70b", 8192, 8192),
("replicate/meta/llama-3-70b-instruct", 8192, 8192),
("replicate/meta/llama-3-8b", 8086, 8086),
("replicate/meta/llama-3-8b-instruct", 8086, 8086),
("replicate/mistralai/mistral-7b-instruct-v0.2", 4096, 4096),
("replicate/mistralai/mistral-7b-v0.1", 4096, 4096),
("replicate/mistralai/mixtral-8x7b-instruct-v0.1", 4096, 4096),
("sagemaker/meta-textgeneration-llama-2-13b", 4096, 4096),
("sagemaker/meta-textgeneration-llama-2-13b-f", 4096, 4096),
("sagemaker/meta-textgeneration-llama-2-70b", 4096, 4096),
("sagemaker/meta-textgeneration-llama-2-70b-b-f", 4096, 4096),
("sagemaker/meta-textgeneration-llama-2-7b", 4096, 4096),
("sagemaker/meta-textgeneration-llama-2-7b-f", 4096, 4096),
("sambanova/deepseek-r1", 32768, 32768),
("sambanova/deepseek-r1-distill-llama-70b", 131072, 131072),
("sambanova/deepseek-v3-0324", 32768, 32768),
("sambanova/deepseek-v3.1", 32768, 32768),
("sambanova/gpt-oss-120b", 131072, 131072),
(
"sambanova/llama-4-maverick-17b-128e-instruct",
131072,
131072,
),
("sambanova/llama-4-scout-17b-16e-instruct", 8192, 8192),
("sambanova/meta-llama-3.1-405b-instruct", 16384, 16384),
("sambanova/meta-llama-3.1-8b-instruct", 16384, 16384),
("sambanova/meta-llama-3.2-1b-instruct", 16384, 16384),
("sambanova/meta-llama-3.2-3b-instruct", 4096, 4096),
("sambanova/meta-llama-3.3-70b-instruct", 131072, 131072),
("sambanova/meta-llama-guard-3-8b", 16384, 16384),
("sambanova/qwen2-audio-7b-instruct", 4096, 4096),
("sambanova/qwen3-32b", 8192, 8192),
("sambanova/qwq-32b", 16384, 16384),
("sarvam/sarvam-m", 8192, 32000),
("snowflake/claude-3-5-sonnet", 18000, 8192),
("snowflake/deepseek-r1", 32768, 8192),
("snowflake/gemma-7b", 8000, 8192),
("snowflake/jamba-1.5-large", 256000, 8192),
("snowflake/jamba-1.5-mini", 256000, 8192),
("snowflake/jamba-instruct", 256000, 8192),
("snowflake/llama2-70b-chat", 4096, 8192),
("snowflake/llama3-70b", 8000, 8192),
("snowflake/llama3-8b", 8000, 8192),
("snowflake/llama3.1-405b", 128000, 8192),
("snowflake/llama3.1-70b", 128000, 8192),
("snowflake/llama3.1-8b", 128000, 8192),
("snowflake/llama3.2-1b", 128000, 8192),
("snowflake/llama3.2-3b", 128000, 8192),
("snowflake/llama3.3-70b", 128000, 8192),
("snowflake/mistral-7b", 32000, 8192),
("snowflake/mistral-large", 32000, 8192),
("snowflake/mistral-large2", 128000, 8192),
("snowflake/mixtral-8x7b", 32000, 8192),
("snowflake/reka-core", 32000, 8192),
("snowflake/reka-flash", 100000, 8192),
("snowflake/snowflake-arctic", 4096, 8192),
("snowflake/snowflake-llama-3.1-405b", 8000, 8192),
("snowflake/snowflake-llama-3.3-70b", 8000, 8192),
("text-completion-codestral/codestral-2405", 32000, 8191),
("text-completion-codestral/codestral-latest", 32000, 8191),
("text-unicorn", 8192, 1024),
("text-unicorn@001", 8192, 1024),
("together-ai-8.1b-21b", 1000, 1000),
("together_ai/deepseek-ai/deepseek-r1", 128000, 20480),
("together_ai/deepseek-ai/deepseek-v3", 65536, 8192),
("together_ai/deepseek-ai/deepseek-v3.1", 128000, 128000),
("together_ai/moonshotai/kimi-k2.5", 256000, 256000),
("together_ai/openai/gpt-oss-120b", 131072, 131072),
("together_ai/zai-org/glm-4.6", 200000, 200000),
("together_ai/zai-org/glm-4.7", 200000, 200000),
(
"us-gov.anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
64000,
),
("us.amazon.nova-2-lite-v1:0", 1000000, 64000),
("us.amazon.nova-2-pro-preview-20251202-v1:0", 1000000, 64000),
("us.amazon.nova-lite-v1:0", 300000, 10000),
("us.amazon.nova-micro-v1:0", 128000, 10000),
("us.amazon.nova-premier-v1:0", 1000000, 10000),
("us.amazon.nova-pro-v1:0", 300000, 10000),
("us.anthropic.claude-3-5-haiku-20241022-v1:0", 200000, 8192),
("us.anthropic.claude-3-5-sonnet-20240620-v1:0", 200000, 4096),
("us.anthropic.claude-3-5-sonnet-20241022-v2:0", 200000, 8192),
("us.anthropic.claude-3-7-sonnet-20250219-v1:0", 200000, 8192),
("us.anthropic.claude-3-haiku-20240307-v1:0", 200000, 4096),
("us.anthropic.claude-3-opus-20240229-v1:0", 200000, 4096),
("us.anthropic.claude-3-sonnet-20240229-v1:0", 200000, 4096),
("us.anthropic.claude-haiku-4-5-20251001-v1:0", 200000, 64000),
("us.anthropic.claude-opus-4-1-20250805-v1:0", 200000, 32000),
("us.anthropic.claude-opus-4-20250514-v1:0", 200000, 32000),
("us.anthropic.claude-opus-4-5-20251101-v1:0", 200000, 64000),
("us.anthropic.claude-opus-4-6-v1", 1000000, 128000),
("us.anthropic.claude-opus-4-7", 1000000, 128000),
("us.anthropic.claude-sonnet-4-20250514-v1:0", 1000000, 64000),
(
"us.anthropic.claude-sonnet-4-5-20250929-v1:0",
200000,
64000,
),
("us.anthropic.claude-sonnet-4-6", 1000000, 64000),
("us.deepseek.r1-v1:0", 128000, 4096),
("us.deepseek.v3.2", 163840, 163840),
("us.meta.llama3-1-405b-instruct-v1:0", 128000, 4096),
("us.meta.llama3-1-70b-instruct-v1:0", 128000, 2048),
("us.meta.llama3-1-8b-instruct-v1:0", 128000, 2048),
("us.meta.llama3-2-11b-instruct-v1:0", 128000, 4096),
("us.meta.llama3-2-1b-instruct-v1:0", 128000, 4096),
("us.meta.llama3-2-3b-instruct-v1:0", 128000, 4096),
("us.meta.llama3-2-90b-instruct-v1:0", 128000, 4096),
("us.meta.llama3-3-70b-instruct-v1:0", 128000, 4096),
("us.meta.llama4-maverick-17b-instruct-v1:0", 128000, 4096),
("us.meta.llama4-scout-17b-instruct-v1:0", 128000, 4096),
("us.mistral.pixtral-large-2502-v1:0", 128000, 4096),
("us.writer.palmyra-x4-v1:0", 128000, 8192),
("us.writer.palmyra-x5-v1:0", 1000000, 8192),
("v0/v0-1.0-md", 128000, 128000),
("v0/v0-1.5-lg", 512000, 512000),
("v0/v0-1.5-md", 128000, 128000),
("vercel_ai_gateway/alibaba/qwen-3-14b", 40960, 16384),
("vercel_ai_gateway/alibaba/qwen-3-235b", 40960, 16384),
("vercel_ai_gateway/alibaba/qwen-3-30b", 40960, 16384),
("vercel_ai_gateway/alibaba/qwen-3-32b", 40960, 16384),
("vercel_ai_gateway/alibaba/qwen3-coder", 262144, 66536),
("vercel_ai_gateway/amazon/nova-lite", 300000, 8192),
("vercel_ai_gateway/amazon/nova-micro", 128000, 8192),
("vercel_ai_gateway/amazon/nova-pro", 300000, 8192),
("vercel_ai_gateway/amazon/titan-embed-text-v2", 0, 0),
(
"vercel_ai_gateway/anthropic/claude-3-5-sonnet",
200000,
8192,
),
(
"vercel_ai_gateway/anthropic/claude-3-5-sonnet-20241022",
200000,
8192,
),
(
"vercel_ai_gateway/anthropic/claude-3-7-sonnet",
200000,
64000,
),
("vercel_ai_gateway/anthropic/claude-3-haiku", 200000, 4096),
("vercel_ai_gateway/anthropic/claude-3-opus", 200000, 4096),
("vercel_ai_gateway/anthropic/claude-3.5-haiku", 200000, 8192),
(
"vercel_ai_gateway/anthropic/claude-3.5-sonnet",
200000,
8192,
),
(
"vercel_ai_gateway/anthropic/claude-3.7-sonnet",
200000,
64000,
),
("vercel_ai_gateway/anthropic/claude-4-opus", 200000, 32000),
("vercel_ai_gateway/anthropic/claude-4-sonnet", 200000, 64000),
(
"vercel_ai_gateway/anthropic/claude-haiku-4.5",
200000,
64000,
),
("vercel_ai_gateway/anthropic/claude-opus-4", 200000, 32000),
("vercel_ai_gateway/anthropic/claude-opus-4.1", 200000, 32000),
("vercel_ai_gateway/anthropic/claude-opus-4.5", 200000, 64000),
("vercel_ai_gateway/anthropic/claude-opus-4.6", 200000, 64000),
("vercel_ai_gateway/anthropic/claude-sonnet-4", 200000, 64000),
(
"vercel_ai_gateway/anthropic/claude-sonnet-4.5",
1000000,
64000,
),
("vercel_ai_gateway/cohere/command-a", 256000, 8000),
("vercel_ai_gateway/cohere/command-r", 128000, 4096),
("vercel_ai_gateway/cohere/command-r-plus", 128000, 4096),
("vercel_ai_gateway/cohere/embed-v4.0", 0, 0),
("vercel_ai_gateway/deepseek/deepseek-r1", 128000, 8192),
(
"vercel_ai_gateway/deepseek/deepseek-r1-distill-llama-70b",
131072,
131072,
),
("vercel_ai_gateway/deepseek/deepseek-v3", 128000, 8192),
("vercel_ai_gateway/google/gemini-2.0-flash", 1048576, 8192),
(
"vercel_ai_gateway/google/gemini-2.0-flash-lite",
1048576,
8192,
),
("vercel_ai_gateway/google/gemini-2.5-flash", 1000000, 65536),
("vercel_ai_gateway/google/gemini-2.5-pro", 1048576, 65536),
("vercel_ai_gateway/google/gemma-2-9b", 8192, 8192),
(
"vercel_ai_gateway/inception/mercury-coder-small",
32000,
16384,
),
("vercel_ai_gateway/meta/llama-3-70b", 8192, 8192),
("vercel_ai_gateway/meta/llama-3-8b", 8192, 8192),
("vercel_ai_gateway/meta/llama-3.1-70b", 128000, 8192),
("vercel_ai_gateway/meta/llama-3.1-8b", 131000, 131072),
("vercel_ai_gateway/meta/llama-3.2-11b", 128000, 8192),
("vercel_ai_gateway/meta/llama-3.2-1b", 128000, 8192),
("vercel_ai_gateway/meta/llama-3.2-3b", 128000, 8192),
("vercel_ai_gateway/meta/llama-3.2-90b", 128000, 8192),
("vercel_ai_gateway/meta/llama-3.3-70b", 128000, 8192),
("vercel_ai_gateway/meta/llama-4-maverick", 131072, 8192),
("vercel_ai_gateway/meta/llama-4-scout", 131072, 8192),
("vercel_ai_gateway/mistral/codestral", 256000, 4000),
("vercel_ai_gateway/mistral/codestral-embed", 0, 0),
("vercel_ai_gateway/mistral/devstral-small", 128000, 128000),
("vercel_ai_gateway/mistral/magistral-medium", 128000, 64000),
("vercel_ai_gateway/mistral/magistral-small", 128000, 64000),
("vercel_ai_gateway/mistral/ministral-3b", 128000, 4000),
("vercel_ai_gateway/mistral/ministral-8b", 128000, 4000),
("vercel_ai_gateway/mistral/mistral-embed", 0, 0),
("vercel_ai_gateway/mistral/mistral-large", 32000, 4000),
("vercel_ai_gateway/mistral/mistral-saba-24b", 32768, 32768),
("vercel_ai_gateway/mistral/mistral-small", 32000, 4000),
(
"vercel_ai_gateway/mistral/mixtral-8x22b-instruct",
65536,
2048,
),
("vercel_ai_gateway/mistral/pixtral-12b", 128000, 4000),
("vercel_ai_gateway/mistral/pixtral-large", 128000, 4000),
("vercel_ai_gateway/moonshotai/kimi-k2", 131072, 16384),
("vercel_ai_gateway/morph/morph-v3-fast", 32768, 16384),
("vercel_ai_gateway/morph/morph-v3-large", 32768, 16384),
("vercel_ai_gateway/openai/gpt-3.5-turbo", 16385, 4096),
(
"vercel_ai_gateway/openai/gpt-3.5-turbo-instruct",
8192,
4096,
),
("vercel_ai_gateway/openai/gpt-4-turbo", 128000, 4096),
("vercel_ai_gateway/openai/gpt-4.1", 1047576, 32768),
("vercel_ai_gateway/openai/gpt-4.1-mini", 1047576, 32768),
("vercel_ai_gateway/openai/gpt-4.1-nano", 1047576, 32768),
("vercel_ai_gateway/openai/gpt-4o", 128000, 16384),
("vercel_ai_gateway/openai/gpt-4o-mini", 128000, 16384),
("vercel_ai_gateway/openai/o1", 200000, 100000),
("vercel_ai_gateway/openai/o3", 200000, 100000),
("vercel_ai_gateway/openai/o3-mini", 200000, 100000),
("vercel_ai_gateway/openai/o4-mini", 200000, 100000),
("vercel_ai_gateway/perplexity/sonar", 127000, 8000),
("vercel_ai_gateway/perplexity/sonar-pro", 200000, 8000),
("vercel_ai_gateway/perplexity/sonar-reasoning", 127000, 8000),
(
"vercel_ai_gateway/perplexity/sonar-reasoning-pro",
127000,
8000,
),
("vercel_ai_gateway/vercel/v0-1.0-md", 128000, 32000),
("vercel_ai_gateway/vercel/v0-1.5-md", 128000, 32768),
("vercel_ai_gateway/xai/grok-2", 131072, 4000),
("vercel_ai_gateway/xai/grok-2-vision", 32768, 32768),
("vercel_ai_gateway/xai/grok-3", 131072, 131072),
("vercel_ai_gateway/xai/grok-3-fast", 131072, 131072),
("vercel_ai_gateway/xai/grok-3-mini", 131072, 131072),
("vercel_ai_gateway/xai/grok-3-mini-fast", 131072, 131072),
("vercel_ai_gateway/xai/grok-4", 256000, 256000),
("vercel_ai_gateway/zai/glm-4.5", 131072, 131072),
("vercel_ai_gateway/zai/glm-4.5-air", 128000, 96000),
("vercel_ai_gateway/zai/glm-4.6", 200000, 200000),
("vertex_ai/claude-3-5-haiku", 200000, 8192),
("vertex_ai/claude-3-5-haiku@20241022", 200000, 8192),
("vertex_ai/claude-3-5-sonnet", 200000, 8192),
("vertex_ai/claude-3-5-sonnet@20240620", 200000, 8192),
("vertex_ai/claude-3-7-sonnet@20250219", 200000, 8192),
("vertex_ai/claude-3-haiku", 200000, 4096),
("vertex_ai/claude-3-haiku@20240307", 200000, 4096),
("vertex_ai/claude-3-opus", 200000, 4096),
("vertex_ai/claude-3-opus@20240229", 200000, 4096),
("vertex_ai/claude-3-sonnet", 200000, 4096),
("vertex_ai/claude-3-sonnet@20240229", 200000, 4096),
("vertex_ai/claude-haiku-4-5", 200000, 8192),
("vertex_ai/claude-haiku-4-5@20251001", 200000, 8192),
("vertex_ai/claude-opus-4", 200000, 32000),
("vertex_ai/claude-opus-4-1", 200000, 32000),
("vertex_ai/claude-opus-4-1@20250805", 200000, 32000),
("vertex_ai/claude-opus-4-5", 200000, 64000),
("vertex_ai/claude-opus-4-5@20251101", 200000, 64000),
("vertex_ai/claude-opus-4-6", 1000000, 128000),
("vertex_ai/claude-opus-4-6@default", 1000000, 128000),
("vertex_ai/claude-opus-4-7", 1000000, 128000),
("vertex_ai/claude-opus-4-7@default", 1000000, 128000),
("vertex_ai/claude-opus-4@20250514", 200000, 32000),
("vertex_ai/claude-sonnet-4", 1000000, 64000),
("vertex_ai/claude-sonnet-4-5", 200000, 64000),
("vertex_ai/claude-sonnet-4-5@20250929", 200000, 64000),
("vertex_ai/claude-sonnet-4-6", 1000000, 64000),
("vertex_ai/claude-sonnet-4-6@default", 1000000, 64000),
("vertex_ai/claude-sonnet-4@20250514", 1000000, 64000),
("vertex_ai/codestral-2", 128000, 128000),
("vertex_ai/codestral-2501", 128000, 128000),
("vertex_ai/codestral-2@001", 128000, 128000),
("vertex_ai/codestral@2405", 128000, 128000),
("vertex_ai/codestral@latest", 128000, 128000),
("vertex_ai/deepseek-ai/deepseek-r1-0528-maas", 65336, 8192),
("vertex_ai/deepseek-ai/deepseek-v3.1-maas", 163840, 32768),
("vertex_ai/deepseek-ai/deepseek-v3.2-maas", 163840, 32768),
("vertex_ai/gemini-3-flash-preview", 1048576, 65535),
("vertex_ai/gemini-3-pro-preview", 1048576, 65535),
("vertex_ai/gemini-3.1-flash-lite-preview", 1048576, 65536),
("vertex_ai/gemini-3.1-pro-preview", 1048576, 65536),
(
"vertex_ai/gemini-3.1-pro-preview-customtools",
1048576,
65536,
),
("vertex_ai/jamba-1.5", 256000, 256000),
("vertex_ai/jamba-1.5-large", 256000, 256000),
("vertex_ai/jamba-1.5-large@001", 256000, 256000),
("vertex_ai/jamba-1.5-mini", 256000, 256000),
("vertex_ai/jamba-1.5-mini@001", 256000, 256000),
("vertex_ai/meta/llama-3.1-405b-instruct-maas", 128000, 2048),
("vertex_ai/meta/llama-3.1-70b-instruct-maas", 128000, 2048),
("vertex_ai/meta/llama-3.1-8b-instruct-maas", 128000, 2048),
(
"vertex_ai/meta/llama-3.2-90b-vision-instruct-maas",
128000,
2048,
),
(
"vertex_ai/meta/llama-4-maverick-17b-128e-instruct-maas",
1000000,
1000000,
),
(
"vertex_ai/meta/llama-4-maverick-17b-16e-instruct-maas",
1000000,
1000000,
),
(
"vertex_ai/meta/llama-4-scout-17b-128e-instruct-maas",
10000000,
10000000,
),
(
"vertex_ai/meta/llama-4-scout-17b-16e-instruct-maas",
10000000,
10000000,
),
("vertex_ai/meta/llama3-405b-instruct-maas", 32000, 32000),
("vertex_ai/meta/llama3-70b-instruct-maas", 32000, 32000),
("vertex_ai/meta/llama3-8b-instruct-maas", 32000, 32000),
("vertex_ai/minimaxai/minimax-m2-maas", 196608, 196608),
("vertex_ai/mistral-large-2411", 128000, 8191),
("vertex_ai/mistral-large@2407", 128000, 8191),
("vertex_ai/mistral-large@2411-001", 128000, 8191),
("vertex_ai/mistral-large@latest", 128000, 8191),
("vertex_ai/mistral-medium-3", 128000, 8191),
("vertex_ai/mistral-medium-3@001", 128000, 8191),
("vertex_ai/mistral-nemo@2407", 128000, 128000),
("vertex_ai/mistral-nemo@latest", 128000, 128000),
("vertex_ai/mistral-small-2503", 128000, 128000),
("vertex_ai/mistral-small-2503@001", 32000, 8191),
("vertex_ai/mistralai/codestral-2", 128000, 128000),
("vertex_ai/mistralai/codestral-2@001", 128000, 128000),
("vertex_ai/mistralai/mistral-medium-3", 128000, 8191),
("vertex_ai/mistralai/mistral-medium-3@001", 128000, 8191),
("vertex_ai/moonshotai/kimi-k2-thinking-maas", 256000, 256000),
("vertex_ai/openai/gpt-oss-120b-maas", 131072, 32768),
("vertex_ai/openai/gpt-oss-20b-maas", 131072, 32768),
(
"vertex_ai/qwen/qwen3-235b-a22b-instruct-2507-maas",
262144,
16384,
),
(
"vertex_ai/qwen/qwen3-coder-480b-a35b-instruct-maas",
262144,
32768,
),
(
"vertex_ai/qwen/qwen3-next-80b-a3b-instruct-maas",
262144,
262144,
),
(
"vertex_ai/qwen/qwen3-next-80b-a3b-thinking-maas",
262144,
262144,
),
("vertex_ai/zai-org/glm-4.7-maas", 200000, 128000),
("vertex_ai/zai-org/glm-5-maas", 200000, 128000),
(
"volcengine/doubao-seed-2-0-code-preview-260215",
256000,
128000,
),
("volcengine/doubao-seed-2-0-lite-260215", 256000, 128000),
("volcengine/doubao-seed-2-0-mini-260215", 256000, 128000),
("volcengine/doubao-seed-2-0-pro-260215", 256000, 128000),
("wandb/deepseek-ai/deepseek-r1-0528", 161000, 161000),
("wandb/deepseek-ai/deepseek-v3-0324", 161000, 161000),
("wandb/deepseek-ai/deepseek-v3.1", 128000, 128000),
("wandb/meta-llama/llama-3.1-8b-instruct", 128000, 128000),
("wandb/meta-llama/llama-3.3-70b-instruct", 128000, 128000),
(
"wandb/meta-llama/llama-4-scout-17b-16e-instruct",
64000,
64000,
),
("wandb/microsoft/phi-4-mini-instruct", 128000, 128000),
("wandb/minimaxai/minimax-m2.5", 197000, 197000),
("wandb/moonshotai/kimi-k2-instruct", 128000, 128000),
("wandb/moonshotai/kimi-k2.5", 262144, 262144),
("wandb/openai/gpt-oss-120b", 131072, 131072),
("wandb/openai/gpt-oss-20b", 131072, 131072),
("wandb/qwen/qwen3-235b-a22b-instruct-2507", 262144, 262144),
("wandb/qwen/qwen3-235b-a22b-thinking-2507", 262144, 262144),
("wandb/qwen/qwen3-coder-480b-a35b-instruct", 262144, 262144),
("wandb/zai-org/glm-4.5", 131072, 131072),
("watsonx/bigscience/mt0-xxl-13b", 8192, 8192),
("watsonx/core42/jais-13b-chat", 8192, 8192),
("watsonx/google/flan-t5-xl-3b", 8192, 8192),
("watsonx/ibm/granite-13b-chat-v2", 8192, 8192),
("watsonx/ibm/granite-13b-instruct-v2", 8192, 8192),
("watsonx/ibm/granite-3-3-8b-instruct", 8192, 8192),
("watsonx/ibm/granite-3-8b-instruct", 8192, 1024),
("watsonx/ibm/granite-4-h-small", 20480, 20480),
("watsonx/ibm/granite-guardian-3-2-2b", 8192, 8192),
("watsonx/ibm/granite-guardian-3-3-8b", 8192, 8192),
("watsonx/ibm/granite-ttm-1024-96-r2", 512, 512),
("watsonx/ibm/granite-ttm-1536-96-r2", 512, 512),
("watsonx/ibm/granite-ttm-512-96-r2", 512, 512),
("watsonx/ibm/granite-vision-3-2-2b", 8192, 8192),
(
"watsonx/meta-llama/llama-3-2-11b-vision-instruct",
128000,
128000,
),
("watsonx/meta-llama/llama-3-2-1b-instruct", 128000, 128000),
("watsonx/meta-llama/llama-3-2-3b-instruct", 128000, 128000),
(
"watsonx/meta-llama/llama-3-2-90b-vision-instruct",
128000,
128000,
),
("watsonx/meta-llama/llama-3-3-70b-instruct", 128000, 128000),
("watsonx/meta-llama/llama-4-maverick-17b", 128000, 128000),
(
"watsonx/meta-llama/llama-guard-3-11b-vision",
128000,
128000,
),
("watsonx/mistralai/mistral-large", 131072, 16384),
("watsonx/mistralai/mistral-medium-2505", 128000, 128000),
("watsonx/mistralai/mistral-small-2503", 32000, 32000),
(
"watsonx/mistralai/mistral-small-3-1-24b-instruct-2503",
32000,
32000,
),
("watsonx/mistralai/pixtral-12b-2409", 128000, 128000),
("watsonx/openai/gpt-oss-120b", 8192, 8192),
("watsonx/sdaia/allam-1-13b-instruct", 8192, 8192),
("writer.palmyra-x4-v1:0", 128000, 8192),
("writer.palmyra-x5-v1:0", 1000000, 8192),
("xai/grok-2", 131072, 131072),
("xai/grok-2-1212", 131072, 131072),
("xai/grok-2-latest", 131072, 131072),
("xai/grok-2-vision", 32768, 32768),
("xai/grok-2-vision-1212", 32768, 32768),
("xai/grok-2-vision-latest", 32768, 32768),
("xai/grok-3", 131072, 131072),
("xai/grok-3-beta", 131072, 131072),
("xai/grok-3-fast-beta", 131072, 131072),
("xai/grok-3-fast-latest", 131072, 131072),
("xai/grok-3-latest", 131072, 131072),
("xai/grok-3-mini", 131072, 131072),
("xai/grok-3-mini-beta", 131072, 131072),
("xai/grok-3-mini-fast", 131072, 131072),
("xai/grok-3-mini-fast-beta", 131072, 131072),
("xai/grok-3-mini-fast-latest", 131072, 131072),
("xai/grok-3-mini-latest", 131072, 131072),
("xai/grok-4", 256000, 256000),
("xai/grok-4-0709", 256000, 256000),
("xai/grok-4-1-fast", 2000000, 2000000),
("xai/grok-4-1-fast-non-reasoning", 2000000, 2000000),
("xai/grok-4-1-fast-non-reasoning-latest", 2000000, 2000000),
("xai/grok-4-1-fast-reasoning", 2000000, 2000000),
("xai/grok-4-1-fast-reasoning-latest", 2000000, 2000000),
("xai/grok-4-fast-non-reasoning", 2000000, 2000000),
("xai/grok-4-fast-reasoning", 2000000, 2000000),
("xai/grok-4-latest", 256000, 256000),
("xai/grok-4.20-0309-reasoning", 2000000, 2000000),
("xai/grok-4.20-beta-0309-non-reasoning", 2000000, 2000000),
("xai/grok-4.20-beta-0309-reasoning", 2000000, 2000000),
("xai/grok-4.20-multi-agent-beta-0309", 2000000, 2000000),
("xai/grok-beta", 131072, 131072),
("xai/grok-code-fast", 256000, 256000),
("xai/grok-code-fast-1", 256000, 256000),
("xai/grok-code-fast-1-0825", 256000, 256000),
("xai/grok-vision-beta", 8192, 8192),
("zai.glm-4.7", 200000, 128000),
("zai.glm-4.7-flash", 200000, 128000),
("zai.glm-5", 200000, 128000),
("zai/glm-4-32b-0414-128k", 128000, 32000),
("zai/glm-4.5", 128000, 32000),
("zai/glm-4.5-air", 128000, 32000),
("zai/glm-4.5-airx", 128000, 32000),
("zai/glm-4.5-flash", 128000, 32000),
("zai/glm-4.5-x", 128000, 32000),
("zai/glm-4.5v", 128000, 32000),
("zai/glm-4.6", 200000, 128000),
("zai/glm-4.7", 200000, 128000),
("zai/glm-5", 200000, 128000),
("zai/glm-5-code", 200000, 128000),
];
#[cfg(test)]
mod tests {
use super::{
LITELLM_MODEL_CAPACITIES, ModelCapacity, catalog_model_capacity,
conservative_model_capacity,
};
#[test]
fn model_capacity_catalog_keys_are_sorted_unique_and_non_empty() {
for window in LITELLM_MODEL_CAPACITIES.windows(2) {
assert!(
window[0].0 < window[1].0,
"model catalog must be strictly sorted for binary search: {:?} before {:?}",
window[0],
window[1]
);
}
for (model_id, _, _) in LITELLM_MODEL_CAPACITIES {
assert!(!model_id.trim().is_empty(), "model id must not be empty");
}
}
#[test]
fn model_capacity_catalog_exact_lookup_normalizes_case_and_space() {
assert_eq!(
catalog_model_capacity(" GPT-4.1 "),
Some(ModelCapacity {
context_window_tokens: 1_047_576,
max_completion_tokens: 32_768,
})
);
}
#[test]
fn model_capacity_catalog_unknown_and_similar_models_do_not_match() {
assert_eq!(catalog_model_capacity("unknown-local-model"), None);
assert_eq!(catalog_model_capacity("gpt-4.1-custom"), None);
assert_eq!(conservative_model_capacity().context_window_tokens, 32_768);
}
}