DeepSeek V3.1 (Non-reasoning)
Open weightsOpen weightsTextDirect answers
DeepSeek V-series family model by DeepSeek — non-reasoning.
About the DeepSeek V-series family
DeepSeek's V2–V3.2 open MoE line (671B total/37B active for V3), including Coder variants, the Terminus and Speciale editions, and V3.2's sparse-attention experiments — each iteration cutting inference cost further.
All DeepSeek V-series modelsHow this one is configured
This configuration runs in non-reasoning mode (direct answers, faster and cheaper).
Who makes it
DeepSeekChinaHangzhou lab famous for frontier-level open models at rock-bottom cost.More from DeepSeek V-series
- DeepSeek Coder V2 Lite Instruct
- DeepSeek LLM 67B Chat (V1)
- DeepSeek V3 (Dec '24)
- DeepSeek V3 0324
- DeepSeek V3.1 (Reasoning)
- DeepSeek V3.1 Terminus (Non-reasoning)
- DeepSeek V3.1 Terminus (Reasoning)
- DeepSeek V3.2 (Non-reasoning)
- DeepSeek V3.2 (Reasoning)
- DeepSeek V3.2 Exp (Non-reasoning)
- DeepSeek V3.2 Exp (Reasoning)
- DeepSeek V3.2 Speciale
- DeepSeek-Coder-V2
- DeepSeek-V2-Chat
- DeepSeek-V2.5
- DeepSeek-V2.5 (Dec '24)