Introducing DeepSeek-V3.1: our first step toward the agent era!
- Hybrid inference: Think & Non-Think — one model, two modes
- Faster thinking: DeepSeek-V3.1-Think reaches answers in less time vs. DeepSeek-R1-0528
- Stronger agent skills: Post-training boosts tool use and multi-step agent tasks
Try it now — toggle Think/Non-Think via the "DeepThink" button: chat.deepseek.com
API Update
- deepseek-chat → non-thinking mode
- deepseek-reasoner → thinking mode
- 128K context for both
- Anthropic API format supported: Anthropic API Guide - API Docs
- Strict Function Calling supported in Beta API: Tool Calls Guide - API Docs
- More API resources, smoother API experience
Tools & Agents Upgrades
- Better results on SWE / Terminal-Bench
- Stronger multi-step reasoning for complex search tasks
- Big gains in thinking efficiency
Model Update
- V3.1 Base: 840B tokens continued pretraining for long context extension on top of V3
- Tokenizer & chat template updated — new tokenizer config: tokenizer_config.json
- V3.1 Base Open-source weights: DeepSeek-V3.1-Base - Hugging Face
- V3.1 Open-source weights: DeepSeek-V3.1 - Hugging Face
Pricing Changes
- New pricing starts & off-peak discounts end at Sep 5th, 2025, 16:00 (UTC Time)
- Until then, APIs follow current pricing
- Pricing page: Models & Pricing - API Docs
