DeepSeek R1 Pruned 108B needs about 66.2 GB of memory at Q4_K_M (a 65.5 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 11 quants, from 39.3 GB (Q2_K) to 113.9 GB (Q8_0). Longer context needs more: at 32K tokens the KV cache alone adds about 0.9 GB.
DeepSeek R1 Pruned 108B is a 107B-parameter DeepSeek model in GGUF format, quantized by mradermacher. Its trained context length is 160K tokens. The repository has 11 quant files, from 39.3 GB (Q2_K) to 113.9 GB (Q8_0).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 39.3 GB | 40.1 GB | Too big | Get file → | |
| Q3_K_S | 46.5 GB | 47.2 GB | Too big | Get file → | |
| Q3_K_M | 51.2 GB | 52.0 GB | Too big | Get file → | |
| Q3_K_L | 55.7 GB | 56.4 GB | Too big | Get file → | |
| IQ4_XS | 57.3 GB | 58.0 GB | Too big | Get file → | |
| Q4_K_S | 60.6 GB | 61.3 GB | Too big | Get file → | |
| Q4_K_M | 65.5 GB | 66.2 GB | Too big | Get file → | |
| Q5_K_S | 73.9 GB | 74.6 GB | Too big | Get file → | |
| Q5_K_M | 76.4 GB | 77.1 GB | Too big | Get file → | |
| Q6_K | 88.0 GB | 88.7 GB | Too big | Get file → | |
| Q8_0 | 113.9 GB | 114.6 GB | Too big | Get file → |
What GPU do you need for DeepSeek R1 Pruned 108B?
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: the best quant that fits is Q3_K_S (46.5 GB file, about 47.2 GB in use).
Which DeepSeek R1 Pruned 108B quant should you choose?
Q4_K_M is the usual starting point for DeepSeek R1 Pruned 108B: it balances file size and quality well. Here it is a 65.5 GB file.
The Q8_0 file is 113.9 GB, about 1.7 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
If memory is tight, the smallest file is Q2_K at 39.3 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 R1 Pruned 108B
How much VRAM does DeepSeek R1 Pruned 108B need?
About 66.2 GB at Q4_K_M with an 8K context window. That is the 65.5 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 R1 Pruned 108B on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 40.1 GB, so you would need a larger GPU or CPU offloading.
What is the file size of DeepSeek R1 Pruned 108B GGUF?
It depends on the quant: Q2_K is 39.3 GB, Q3_K_S is 46.5 GB, Q3_K_M is 51.2 GB, Q3_K_L is 55.7 GB, IQ4_XS is 57.3 GB, Q4_K_S is 60.6 GB and 5 more.
What context length does DeepSeek R1 Pruned 108B support?
The GGUF header lists a trained context length of 163,840 tokens (160K). The KV cache grows with the context you use: about 0.2 GB at 8K and 0.9 GB at 32K.
Where can I download DeepSeek R1 Pruned 108B GGUF?
The files are on Hugging Face in the mradermacher/DeepSeek-R1-Pruned-108B-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.