Reasoning models
Reasoning models emit internal chain-of-thought tokens before their final answer. Higher latency, much higher accuracy on hard problems.
9 models in this category. See also the “Best AI reasoning models” ranking with editorial picks and FAQ.
| Model | Provider | Context | Input price / 1M | Tier |
|---|---|---|---|---|
| o1 openai/o1 | OpenAI | 200K Long context (128K+) | Custom | Variable |
| o3-mini openai/o3-mini | OpenAI | 200K Long context (128K+) | Custom | Variable |
| DeepSeek R1 deepseek/deepseek-r1 | DeepSeek | 128K Long context (128K+) | Custom | Variable |
| Claude Sonnet 4.6 anthropic/claude-sonnet-4.6 | Anthropic | 1M Ultra context (1M+) | Custom | Variable |
| DeepSeek Chat (legacy id) deepseek/deepseek-chat | DeepSeek | 1M Ultra context (1M+) | Custom | Variable |
| DeepSeek Reasoner (legacy id) deepseek/deepseek-reasoner | DeepSeek | 1M Ultra context (1M+) | Custom | Variable |
| DeepSeek V4 Flash deepseek/deepseek-v4-flash | DeepSeek | 1M Ultra context (1M+) | Custom | Variable |
| Claude Opus 4.7 anthropic/claude-opus-4.7 | Anthropic | 1M Ultra context (1M+) | Custom | Variable |
| Claude 3 Opus anthropic/claude-3-opus | Anthropic | 200K Long context (128K+) | Custom | Variable |