Expand description
Pre-configured model definitions
This module contains model structs for various LLM providers.
Each model implements the Model trait and provider-specific traits
like BedrockModel or AnthropicModel.
Models are organized by vendor:
claude- Anthropic Claude modelsllama- Meta Llama modelsnova- Amazon Nova modelsmistral- Mistral AI modelscohere- Cohere modelsqwen- Alibaba Qwen modelsgoogle- Google modelsdeepseek- DeepSeek modelskimi- Moonshot Kimi models
Structsยง
- Claude3_
7Sonnet - Claude 3.7 Sonnet - Latest Claude 3.x with improved reasoning
- Claude
Haiku4_ 5 - Claude Haiku 4.5 - Fast, efficient model for high-throughput tasks
- Claude
Opus4 - Claude Opus 4 - High capability reasoning model
- Claude
Opus4_ 1 - Claude Opus 4.1 - Advanced reasoning model
- Claude
Opus4_ 5 - Claude Opus 4.5 - High-capability reasoning and creative writing model
- Claude
Opus4_ 6 - Claude Opus 4.6 - Flagship Claude model with 128K output
- Claude
Sonnet4 - Claude Sonnet 4 - Balanced performance and cost
- Claude
Sonnet4_ 5 - Claude Sonnet 4.5 - Latest Sonnet with improved capabilities
- Cohere
CommandR Plus - Command R+ - Enterprise RAG and multi-step tool use model
- Deep
Seek R1 - DeepSeek R1 - Reasoning-focused model
- Deep
Seek V3_ 1 - DeepSeek V3.1 - General purpose model
- Deep
Seek V3_ 2 - DeepSeek V3.2 - Updated general purpose model
- Gemma3_
4B - Gemma 3 4B - Compact open model from Google
- Gemma3_
12B - Gemma 3 12B - Mid-size open model from Google
- Gemma3_
27B - Gemma 3 27B - Open multimodal model from Google
- Kimi
K2Thinking - Kimi K2 Thinking - Reasoning-enhanced model from Moonshot AI
- Kimi
K2_ 5 - Kimi K2.5 - Next-gen model from Moonshot AI
- Llama3_
1_ 8B - Llama 3.1 8B Instruct - Efficient general purpose model
- Llama3_
1_ 70B - Llama 3.1 70B Instruct - High capability model
- Llama3_
1_ 405B - Llama 3.1 405B Instruct - Largest open-weights model
- Llama3_
2_ 1B - Llama 3.2 1B Instruct - Lightweight model for edge deployment
- Llama3_
2_ 3B - Llama 3.2 3B Instruct - Efficient small model
- Llama3_
2_ 11B - Llama 3.2 11B Instruct - Medium multimodal model
- Llama3_
2_ 90B - Llama 3.2 90B Instruct - Large multimodal model
- Llama3_
3_ 70B - Llama 3.3 70B Instruct - Latest Llama 3.x flagship
- Llama4
Maverick17B - Llama 4 Maverick 17B - Larger MoE model with 1M context
- Llama4
Scout17B - Llama 4 Scout 17B - Efficient MoE model with 10M context
- Magistral
Small - Magistral Small - Efficient 24B reasoning model with vision
- Ministral3B
- Ministral 3B - Compact 3B instruction model
- Ministral8B
- Ministral 8B - Efficient 8B instruction model
- Ministral14B
- Ministral 14B - Mid-size 14B instruction model
- Mistral
Large3 - Mistral Large 3 - Flagship 675B MoE model with 41B active parameters
- Nova2
Lite - Nova 2 Lite - Fast reasoning model with extended thinking support
- Nova2
Sonic - Nova 2 Sonic - Next-gen Nova model with 1M context
- Nova
Lite - Nova Lite - Multimodal model for image, video, and text
- Nova
Micro - Nova Micro - Lightweight, text-only model for simple tasks
- Nova
Premier - Nova Premier - Highest capability Nova model with 1M context
- NovaPro
- Nova Pro - Balanced multimodal model
- Pixtral
Large - Pixtral Large - Vision-capable large model
- Qwen3
Coder30B - Qwen3 Coder 30B - Compact coding-focused MoE model
- Qwen3
Coder480B - Qwen3 Coder 480B - Large coding-focused MoE model
- Qwen3
Next80B - Qwen3 Next 80B - Next-gen MoE model
- Qwen3V
L235B - Qwen3 VL 235B - Vision-language MoE model
- Qwen3_
32B - Qwen3 32B - Dense 32B model
- Qwen3_
235B - Qwen3 235B - Large MoE model with 22B active parameters
- Voxtral
Mini3B - Voxtral Mini 3B - Speech and text input model
- Voxtral
Small24B - Voxtral Small 24B - Speech and text input model