DeepSeek V3.1
Provider：DeepSeek
DeepSeek-V3.1 is a large hybrid reasoning model (671B parameters, 37B active) that supports both thinking and non-thinking modes via prompt templates. It extends the DeepSeek-V3 base with a two-phase long-context...
Key specs
- Context：163.8K
- Max output：144.9K
- Tokenizer：DeepSeek
- Released：2025-08-21
Token pricing
- Input price：$0.550 / 1M tokens
- Output price：$1.65 / 1M tokens
- Cache read：$0.550 / 1M tokens
- Blended price：$0.825 / 1M tokens
Modalities
Text
Use Cases
- Deep Reasoning & Analysis：Math proofs, logical reasoning, multi-step planning — ideal for research and financial analysis.（≈ $0.013/call）
- Chat & Support：FAQ auto-response, multi-turn dialogue, emotion detection — e-commerce, finance, gov.（≈ $0.0033/call）
When to pick this model
Signals derived from public pricing and spec fields — not a benchmark. Validate with official docs and your own eval.
- Long max output (>=30K tokens): strong fit for deep report / essay / document-drafting workflows.
FAQ
- What is the token pricing of DeepSeek V3.1? Input price $0.550 / 1M tokens, output price $1.65 / 1M tokens, blended about $0.825 / 1M tokens. Refer to DeepSeek’s official page for the exact rate.
- What context window does DeepSeek V3.1 support? Context window is 163.8K, max output about 144.9K, suitable for long documents, multi-turn chat and complex reasoning.
- What scenarios is DeepSeek V3.1 best for? Based on its capability and pricing, DeepSeek V3.1 fits best: Deep Reasoning & Analysis, Chat & Support. See the Use Cases section for details and cost estimates.
- What input/output modalities does DeepSeek V3.1 support? Supports Text modality.
- Is DeepSeek V3.1 a free model? DeepSeek V3.1 is billed per token, not a free model.
- Which provider offers DeepSeek V3.1? DeepSeek V3.1 is offered by DeepSeek.
More models from DeepSeek
Full info：DeepSeek V3.1