DeepSeek V3.2 — local AI model by DeepSeek
Written by Jakub Rusinowski · Last updated
DeepSeek's updated 685B MoE model delivering approximately 90% of GPT-5 quality at 1/50th the price. Significant performance jump over V3 — AIME 2026 at 91.67% in thinking mode. MIT-licensed and widely adopted. Input tokens at ~$0.28/1M make it one of the most cost-effective frontier models via API. Self-hostable on multi-GPU clusters (~370 GB at Q4). DEPRECATED: DeepSeek V3.2 is being retired (reported 2026-07-24) in favour of DeepSeek V4 (see the "DeepSeek V4" family) — kept here for users still running it.
Variants
The smallest DeepSeek V3.2 variant needs about 414 GB of VRAM at Q4_K_M — quantized weights plus framework overhead, before any KV cache.
| Model | VRAM at Q4 | VRAM | Context | Run it |
|---|---|---|---|---|
| DeepSeek V3.2 671B → 671B/37B active | ~414.4 GB | 128,000 | ollama run deepseek-v3.2:671b-q4 | |
| DeepSeek V3.2 685B 685B | ~414.4 GB | 128,000 | ollama run deepseek-v3-2 |
Memory is quantized weights plus overhead at Q4_K_M, from the same engine as the GPU & VRAM checker.
How to run DeepSeek V3.2 locally
Install Ollama, then pull the tag.
ollama run deepseek-v3.2:671b-q4Pick a size above for its own VRAM figure, speed estimate and install command.
Licence
Commercial use permitted. No usage restrictions beyond attribution.
Applies to: DeepSeek V3.2 671B, DeepSeek V3.2 685BCan I run DeepSeek V3.2 on my GPU?
DeepSeek V3.2 — frequently asked questions
How much VRAM does DeepSeek V3.2 need?
DeepSeek V3.2 needs about 414 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: DeepSeek V3.2 671B (414 GB, Q4_K_M); DeepSeek V3.2 685B (414 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Can I run DeepSeek V3.2 on an RTX 4090 (24 GB)?
DeepSeek V3.2's smallest variant needs about 414 GB, which exceeds a single RTX 4090 (24 GB). Use multiple GPUs, a higher-VRAM card, or Apple Silicon with large unified memory.
What quantization should I use for DeepSeek V3.2?
Q4_K_M is the best balance of quality and VRAM for DeepSeek V3.2 in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
How do I run DeepSeek V3.2 with Ollama?
Install Ollama, then run: ollama run deepseek-v3.2:671b-q4. This downloads DeepSeek V3.2 and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.