DeepSeek-V3.1
DeepSeek-V3.1 is an open-weights language model from DeepSeek, released August 21, 2025. It has 685B parameters (mixture-of-experts, 51B active). The context window is 160K tokens. On the Epoch Capabilities Index it scores 140.8, rank 120 of 256 rated models.
DeepSeek-V3.1 is a hybrid model that supports both thinking mode and non-thinking mode. Compared to the previous version, this upgrade brings improvements in multiple aspects: Hybrid thinking mode: One model supports both thinking mode and non-thinking mode by changing the chat template.
Check it against your own PC →
Can you run DeepSeek-V3.1 locally?
Even with 32 GB of VRAM and 32 GB of RAM it is too large to run at a usable speed on a single consumer GPU. Estimates assume 32 GB of DDR5 RAM, a 16K context and llama.cpp-style GGUF quantization. Enter your exact hardware in the live app.
| GPU memory | Verdict | Best quant | VRAM | RAM | tok/s |
|---|---|---|---|---|---|
| 6 GB VRAM | Too large | – | – | – | – |
| 8 GB VRAM | Too large | – | – | – | – |
| 12 GB VRAM | Too large | – | – | – | – |
| 16 GB VRAM | Too large | – | – | – | – |
| 24 GB VRAM | Too large | – | – | – | – |
| 32 GB VRAM | Too large | – | – | – | – |
Ready-made quantizations on Hugging Face: bartowski/deepseek-ai_DeepSeek-V3.1-Base-GGUF · deepseek-ai/DeepSeek-V3.1 · deepseek-ai/DeepSeek-V3.1-Terminus · gamm0000/DeepSeek-V3.1-Terminus · unsloth/DeepSeek-V3.1-Base · sparrow18/DeepSeek-V3.1-Terminus
Benchmark results
| Benchmark (Epoch AI) | Score |
|---|---|
| SimpleBench | 28.0% |
Links
Announcement · Hugging Face: deepseek-ai/DeepSeek-V3.1-Base · Epoch AI benchmarks
Models with a similar score
Claude Opus 4.1 · DeepSeek-V3.2-Exp · Grok 4 Fast · Gemma 4 31B IT · Qwen 3.6 35B-A3B · Qwen3.5-35B-A3B · Gemini 2.5 Pro (Mar 2025) · Qwen 3.5 Flash (hosted 35B-A3B)
More from DeepSeek
DeepSeek V4.1 Flash 1017 · DeepSeek V4 Flash Vision Exp · DeepSeek V4.1 Flash · DeepSeek V4 Pro 0813 · DeepSeek V4 Flash 0731 · DeepSeek V4 Pro DSpark · DeepSeek V4 Flash DSpark · DeepSeek-V4-Pro