Local LLM model page

DeepSeek V3.2 (37B/671B MoE)

DeepSeek's massive MoE flagship. 37B active out of 671B total. Exceptional coding, reasoning and general capabilities. Ranks #6 on global usage leaderboards with 29B monthly tokens. MIT licensed.

Parameters
37B (671B MoE)
Minimum RAM
48 GB
Model size
40 GB
Quantization
Q4_K_M

Can DeepSeek V3.2 (37B/671B MoE) run locally?

DeepSeek V3.2 (37B/671B MoE) is best suited for high-end workstations with 64 GB RAM. LocalClaw recommends Q4_K_M as the default quantization, with at least 48 GB RAM.

Search term for LM Studio or compatible runtimes: deepseek-v3.2

Hugging Face repository: deepseek-ai/DeepSeek-V3.2-GGUF

chatcodereasoningpowerqualitygeneral

Strengths

  • Latest DeepSeek MoE flagship
  • Ranks #6 globally with 29B monthly tokens
  • MIT license
  • Exceptional coding and reasoning

Limitations

  • Requires 48GB+ RAM
  • Server-grade recommended
  • Complex setup

Best use cases

  • Enterprise AI
  • Research
  • Complex coding
  • Reasoning at scale

Benchmarks

Speed: 3/10

Quality: 10/10

Coding: 10/10

Reasoning: 10/10

Technical details

Developer: DeepSeek AI

License: MIT

Context window: 131,072 tokens

Architecture: Mixture of Experts (MoE) — 671B total, ~37B active

Released: 2025-12