Best Local DeepSeek Models

The best local LLMs — open-weight models small enough to run on your own hardware with Ollama, llama.cpp or vLLM — ranked by intelligence. Frontier-scale open models (400B+) are excluded: downloadable isn't the same as runnable. Every model here is on Hugging Face.

1. 1
   deepseek-v4-pro

   ReasoningToolsJSON44.3 intel · $0.435/M · 1.0M ctx
   44.3

   Intelligence
2. 2
   deepseek-v4-flash

   ReasoningToolsJSON40.3 intel · $0.098/M · 1.0M ctx
   40.3

   Intelligence
3. 3
   deepseek-v3.2

   ReasoningToolsJSON24.7 intel · $0.269/M · 164K ctx
   24.7

   Intelligence
4. 4
   deepseek-v3.1-terminus

   ReasoningToolsJSON21.4 intel · $0.270/M · 164K ctx
   21.4

   Intelligence
5. 5
   deepseek-r1

   ReasoningToolsJSON20.1 intel · $0.700/M · 164K ctx
   20.1

   Intelligence