Strong coding and instruction-following for agents.
Model details →Claude 3.7 Sonnet vs DeepSeek Reasoner (legacy id)
DeepSeek Reasoner (legacy id) wins on 1 of 6 axes — pricing and capability skew in its favour for most workloads.
Legacy compatibility id mapping to DeepSeek V4 Flash thinking mode (per DeepSeek pricing notes).
Model details →Side-by-side comparison
| Capability | Claude 3.7 Sonnet | DeepSeek Reasoner (legacy id) | Winner |
|---|---|---|---|
| Context window Maximum number of input tokens the model can attend to in a single request. | 200K | 1M | 🏆 DeepSeek Reasoner (legacy id) |
| Input price (per 1M) Cost per million input tokens billed by the provider. | Custom | Custom | — |
| Output price (per 1M) Cost per million output tokens billed by the provider. | Custom | Custom | — |
| Tool / function calling First-class support for emitting structured tool calls. | — | — | Tie |
| Vision input Accepts image inputs alongside text. | — | — | Tie |
| Reasoning mode Internal chain-of-thought / extended-thinking support. | — | — | Tie |
Frequently asked questions
Is Claude 3.7 Sonnet better than DeepSeek Reasoner (legacy id)?
DeepSeek Reasoner (legacy id) wins on 1 of 6 axes — pricing and capability skew in its favour for most workloads.
What's the price difference between Claude 3.7 Sonnet and DeepSeek Reasoner (legacy id)?
Input: Custom vs Custom per 1M tokens. Output: Custom vs Custom per 1M tokens.
What context windows do Claude 3.7 Sonnet and DeepSeek Reasoner (legacy id) support?
Claude 3.7 Sonnet supports up to 200K tokens. DeepSeek Reasoner (legacy id) supports up to 1M tokens.
Do both Claude 3.7 Sonnet and DeepSeek Reasoner (legacy id) support tool calling?
Claude 3.7 Sonnet: not advertised. DeepSeek Reasoner (legacy id): not advertised.