DeepSeek: deepseek-v3.1 API
A hybrid-inference flagship MoE model (671B total, 37B active). It bridges the gap between speed and logic by supporting both Thinking and Non-Thinking modes in a single model, featuring enhanced tool-calling and a 128k context window.
- Context window: 128,000 tokens
- Max output: 32,768 tokens
- Input: text
- Output: text
- Reasoning: Supported
- Tool calling: Supported
- File input: Supported
- Released: 2025-08-21
- Knowledge cutoff: 2025-05-31
Frequently Asked Questions
What is the context window of deepseek-v3.1?
deepseek-v3.1 supports a context window of up to 128,000 tokens.
Does deepseek-v3.1 support function calling?
Yes. deepseek-v3.1 supports tool / function calling.
Does deepseek-v3.1 support reasoning?
Yes. deepseek-v3.1 is a reasoning-capable model.
What is the knowledge cutoff of deepseek-v3.1?
The knowledge cutoff of deepseek-v3.1 is 2025-05-31.