Does It Run Local?

← All comparisons

MODEL COMPARISON

DeepSeek-V3.2 vs DeepSeek-V3

DeepSeek-V3.2 and DeepSeek-V3 need about the same memory (within 1 GB): about 488 GB and about 487 GB. DeepSeek-V3.2 has no runtime listed as in release. DeepSeek-V3 has one runtime in release (Ollama).

SIDE BY SIDE

CompareDeepSeek-V3.2DeepSeek-V3
Memory neededabout 488 GB · Up to 512 GBabout 487 GB · Up to 512 GB
Recommended quantQ4_K_M · 405.4 GBQ4_K_M · 404.4 GB
llama.cpp supportArchitecture listedChecked 28 Oct 2026Architecture listedChecked 28 Oct 2026
Ollama supportUnknownVersion v0.40.12 · Checked 28 Oct 2026ReleasedChecked 28 Oct 2026
LM Studio supportUnknownChecked 28 Oct 2026UnknownChecked 28 Oct 2026
MLX / mlx-lm supportUpload availableChecked 28 Oct 2026Upload availableChecked 28 Oct 2026
vLLM supportArchitecture listedChecked 28 Oct 2026Architecture listedChecked 28 Oct 2026
ReleasedDec 1, 2025Dec 25, 2024
LicenseNot recorded in this snapshot — see the model card ↗Not recorded in this snapshot — see the model card ↗
Publisherdeepseek-aideepseek-ai
Downloads6M1.6M
Model pageOpen DeepSeek-V3.2 ↗Open DeepSeek-V3 ↗
Fits these setups
  • Mac 16 GBWon't fit
  • Mac 24 GBWon't fit
  • Mac 32 GBWon't fit
  • Mac 64 GBWon't fit
  • GPU 8 GB + 32 GB RAMWon't fit
  • GPU 12 GB + 32 GB RAMWon't fit
  • GPU 16 GB + 32 GB RAMWon't fit
  • GPU 24 GB + 64 GB RAMWon't fit
  • CPU 32 GBWon't fit
  • CPU 64 GBWon't fit
  • Mac 16 GBWon't fit
  • Mac 24 GBWon't fit
  • Mac 32 GBWon't fit
  • Mac 64 GBWon't fit
  • GPU 8 GB + 32 GB RAMWon't fit
  • GPU 12 GB + 32 GB RAMWon't fit
  • GPU 16 GB + 32 GB RAMWon't fit
  • GPU 24 GB + 64 GB RAMWon't fit
  • CPU 32 GBWon't fit
  • CPU 64 GBWon't fit

Memory needed is the same estimate shown on each model page: the recommended observed file plus 20% headroom and a 1.5 GB runtime allowance. See which setups fit →

These are estimates from observed file sizes, not benchmarks. They don’t account for long contexts or the KV cache, and fitting does not establish runtime compatibility or speed.