Claude Haiku 4.5 supports text, image input and text output with 200K context, suited for coding agent, vision qa, long context.
Claude Opus 4.7 supports text, image input and text output with 200K context, suited for coding agent, vision qa, reasoning.
Claude Sonnet 4.6 supports text, image input and text output with 200K context, suited for coding agent, vision qa, reasoning.
Codestral 25.01 supports text input and text output with 256K context, suited for coding agent, long context, cheap batch.
Cohere Embed v3 (English) supports text input and text, embeddings output with 1K context, suited for cheap batch, multimodal, embedding.
Cohere Rerank v3 (multilingual) supports text input and text, rerank output with 4K context, suited for multimodal, rerank.
Command R supports text input and text output with 128K context, suited for coding agent, cheap batch, tool_use.
Command R+ supports text input and text output with 128K context, suited for coding agent, tool_use.
DeepSeek V4 Flash supports text input and text output with 1M context, suited for coding agent, reasoning, long context.
DeepSeek V4 Pro supports text input and text output with 1M context, suited for coding agent, reasoning, long context.
ERNIE 4.5 Turbo supports text input and text output with 128K context, suited for coding agent, cheap batch, tool_use.
ERNIE X1 Turbo supports text input and text output with 32K context, suited for coding agent, reasoning, cheap batch.
GLM-4 FlashX supports text input and text output with 128K context, suited for coding agent, cheap batch, tool_use.
GLM-4 Plus supports text input and text output with 128K context, suited for coding agent, tool_use.
GLM-4.6 supports text input and text output with 200K context, suited for coding agent, reasoning, long context.
GLM-4V supports text, image input and text output with 8K context, suited for vision qa, multimodal, vision.
GLM-5 supports text, image input and text output with 200K context, suited for coding agent, vision qa, reasoning.
GPT-4o supports text, image input and text output with 128K context, suited for coding agent, vision qa, multimodal.
GPT-4o Mini supports text, image input and text output with 128K context, suited for coding agent, vision qa, cheap batch.
GPT-5 Mini supports text, image input and text output with 200K context, suited for coding agent, vision qa, long context.
GPT-5.3 supports text, image input and text output with 400K context, suited for coding agent, vision qa, reasoning.
GPT-5.3 Codex supports text input and text output with 400K context, suited for coding agent, reasoning, long context.
Gemini 2.5 Flash supports text, image input and text output with 1M context, suited for coding agent, vision qa, long context.
Gemini 2.5 Pro supports text, image input and text output with 2M context, suited for coding agent, vision qa, reasoning.
Gemini 3 Flash supports text, image input and text output with 1M context, suited for coding agent, vision qa, long context.
Gemini 3 Pro supports text, image input and text output with 2M context, suited for coding agent, vision qa, reasoning.
Grok 3 supports text, image input and text output with 128K context, suited for coding agent, vision qa, multimodal.
Grok 4 supports text, image input and text output with 256K context, suited for coding agent, vision qa, reasoning.
Hunyuan Pro supports text input and text output with 32K context, suited for coding agent, tool_use.
Hunyuan Turbo supports text input and text output with 32K context, suited for coding agent, tool_use.
Kimi K2.5 supports text, image input and text output with 262K context, suited for coding agent, vision qa, long context.
Kimi K2.6 supports text, image input and text output with 262K context, suited for coding agent, vision qa, reasoning.
Llama 3.3 70B supports text input and text output with 128K context, suited for coding agent, tool_use.
MiniMax Text 01 supports text input and text output with 1M context, suited for coding agent, long context, cheap batch.
MiniMax abab7-chat supports text input and text output with 246K context, suited for coding agent, long context, tool_use.
Mistral Large 2 (24.11) supports text input and text output with 128K context, suited for coding agent, tool_use.
Mistral Medium 3 supports text input and text output with 128K context, suited for coding agent, cheap batch, tool_use.
Mistral Small 3 supports text input and text output with 128K context, suited for coding agent, cheap batch, tool_use.
Moonshot v1 128k supports text input and text output with 128K context, suited for coding agent, tool_use.
Moonshot v1 32k supports text input and text output with 32K context, suited for coding agent, tool_use.
OpenAI o1 supports text input and text output with 200K context, suited for reasoning, long context, thinking.
Pixtral Large supports text, image input and text output with 131K context, suited for coding agent, vision qa, multimodal.
Qwen Max supports text input and text output with 32K context, suited for coding agent, tool_use.
Qwen Plus supports text input and text output with 131K context, suited for coding agent, cheap batch, tool_use.
Qwen Turbo supports text input and text output with 1M context, suited for coding agent, long context, cheap batch.
Qwen3 Coder supports text input and text output with 256K context, suited for coding agent, long context, cheap batch.
Qwen3 VL supports text, image input and text output with 128K context, suited for coding agent, vision qa, multimodal.
Yi Large supports text input and text output with 33K context, suited for coding agent, tool_use.
Yi Vision supports text, image input and text output with 16K context, suited for vision qa, multimodal, vision.