How much VRAM does DeepSeek V3.2 need?
DeepSeek V3.2 needs 379 GB for its weights at 4-bit, 380 GB with an 8K context and 382 GB with 32K.
- Parameters
- 671B
- Active per token
- 37B
- Longest context
- 160K
- Released
- Dec 2025
- License
- MIT
- Strongest at
- Chat, Coding, Agents, Docs
Memory at each precision
| Precision | Weights | With 8K | With 32K | With 128K |
|---|---|---|---|---|
| As released (FP8) | 642 GB | 644 GB | 645 GB | 653 GB |
| 6-bit | 512 GB | 514 GB | 516 GB | 523 GB |
| 5-bit | 444 GB | 446 GB | 448 GB | 455 GB |
| 4-bit | 379 GB | 380 GB | 382 GB | 389 GB |
| 3-bit | 305 GB | 307 GB | 309 GB | 316 GB |
| 2-bit | 234 GB | 236 GB | 238 GB | 245 GB |
| 1.6-bit | 129 GB | 130 GB | 132 GB | 139 GB |
One conversation, 16-bit context cache, runtime overhead included.
What runs it
| Machine | Best fit | Context | Speed | First reply |
|---|---|---|---|---|
| RTX 3060 12 GB | Too big | |||
| RTX 4060 Ti 16 GB | Too big | |||
| RTX 3090 | Too big | |||
| RTX 4090 | Too big | |||
| RTX 5090 | Too big | |||
| 2× RTX 3090 | Too big | |||
| RTX PRO 6000 Blackwell | Too big | |||
| M4 Max 128 GB | Too big | |||
| M5 Max 128 GB | Too big | |||
| Ryzen AI Max+ 395 128 GB | Too big | |||
| DGX Spark 128 GB | Too big | |||
| M3 Ultra 256 GB | 1.6-bit | 160K | 44 tok/s | 23 s |
| M5 Ultra 512 GB | 3-bit | 160K | 32 tok/s | 7.2 s |
| H100 SXM | Too big | |||
| DGX Station GB300 | 2-bit | 52K | 155 tok/s | 0.7 s |
| 8× H200 NVL | FP8 | 160K | 67 tok/s | 0.3 s |
Best fit is the highest quality, up to 8-bit, that leaves 16K of context. First reply is the wait for a 2K prompt. Speeds are rough, plus or minus 25%.
How to run it
- llama.cpp
llama-server -hf unsloth/DeepSeek-V3.2-GGUF- vLLM
vllm serve deepseek-ai/DeepSeek-V3.2- MLX (Mac)
mlx_lm.chat --model mlx-community/DeepSeek-V3.2-4bit
Questions
How much VRAM does DeepSeek V3.2 need?
DeepSeek V3.2 needs 379 GB for its weights at 4-bit, 380 GB with an 8K context and 382 GB with 32K.
Can DeepSeek V3.2 run on an RTX 4090?
Not on one card. The smallest machine here that runs it is the M3 Ultra 256 GB.