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 training process, reaching up to 128K tokens, and uses FP8 microscaling for efficient inference. Users can control the reasoning behaviour with the `reasoning` `enabled` boolean. [Learn more in our docs](https://ask-coreai.com) The model improves tool use, code generation, and reasoning efficiency, achieving performance comparable to DeepSeek-R1 on difficult benchmarks while responding more quickly. It supports structured tool calling, code agents, and search agents, making it suitable for research, coding, and agentic workflows. It succeeds the [DeepSeek V3-0324](/deepseek/deepseek-chat-v3-0324) model and performs well on a variety of tasks.
With CoreAI, you can start chatting with DeepSeek: DeepSeek V3.1 instantly — no separate subscription needed. CoreAI bundles access to DeepSeek: DeepSeek V3.1 along with 300+ other AI models from DeepSeek and other providers like OpenAI, Anthropic, Google, Meta, and more.
Chat with DeepSeek: DeepSeek V3.1 and 300+ other AI models — all in one app.