Deepseek needs about 182.6 GB of memory at its main file (a 141.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 8 quants, from 62.7 GB (IQ1_S) to 471.6 GB (BF16). Longer context needs more: at 32K tokens the KV cache alone adds about 161.1 GB.
Deepseek is a 236B-parameter DeepSeek model in GGUF format, quantized by nisten. Its trained context length is 160K tokens. The repository has 8 quant files, from 62.7 GB (IQ1_S) to 471.6 GB (BF16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| IQ1_S | 62.7 GB (2 parts) | 103.5 GB | Too big | Get file → | |
| — | 78.7 GB (2 parts) | 119.4 GB | Too big | Get file → | |
| IQ2_XXS | 86.5 GB (2 parts) | 127.3 GB | Too big | Get file → | |
| — | 141.9 GB (4 parts) | 182.6 GB | Too big | Get file → | |
| Q4_0 | 148.8 GB (4 parts) | 189.5 GB | Too big | Get file → | |
| — | 201.0 GB (5 parts) | 241.8 GB | Too big | Get file → | |
| Q8_0 | 250.6 GB (6 parts) | 291.4 GB | Too big | Get file → | |
| BF16 | 471.6 GB (11 parts) | 512.4 GB | Too big | Get file → |
What GPU do you need for Deepseek?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.
- 4 GB VRAM: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which Deepseek quant should you choose?
If memory is tight, the smallest file is IQ1_S at 62.7 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Deepseek
How much VRAM does Deepseek need?
About 182.6 GB at its main file with an 8K context window. That is the 141.9 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.
Can I run Deepseek on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 103.5 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Deepseek GGUF?
It depends on the quant: IQ1_S is 62.7 GB, file is 78.7 GB, IQ2_XXS is 86.5 GB, file is 141.9 GB, Q4_0 is 148.8 GB, file is 201.0 GB and 2 more.
What context length does Deepseek support?
The GGUF header lists a trained context length of 163,840 tokens (160K). The KV cache grows with the context you use: about 40.3 GB at 8K and 161.1 GB at 32K.
Where can I download Deepseek GGUF?
The files are on Hugging Face in the nisten/deepseek-0628-gguf repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does Deepseek use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.