DeepSeek GGUF models
4419 GGUF models in the DeepSeek family, quantized by 1572 uploaders. Sizes run from 1M to 5376B parameters. The longest trained context is 1024K tokens.
About the DeepSeek GGUF models
DeepSeek is represented on ggufmodels.com by 4,419 GGUF models, quantized by 1,572 uploaders on Hugging Face. The most downloaded are DeepSeek V4.1 Flash 1017 GSQ RCO, DeepSeek V4.1 Flash Vision Encoder and Huihui DeepSeek V4 Flash 0731 abliterated. Parameter counts run from 1M to 5376B, and the longest trained context window is 1024K tokens.
A Q4_K_M file of a DeepSeek model is between 0.0 to 954.6 GB, which is the quant most people start with. Most files come from mradermacher (955), roleplaiapp (222) and Thireus (165). The quants you will meet most often are Q4_K_M (2356), Q8_0 (1841), Q5_K_M (1587) and Q2_K (1574), where the number is how many models offer that quant.
Frequently asked questions
Which DeepSeek GGUF should I download?
The most downloaded DeepSeek GGUF here is DeepSeek V4.1 Flash 1017 GSQ RCO. Pick the size that fits your GPU first, then choose a quant: Q4_K_M is the usual starting point, and Q5_K_M or Q6_K give more quality if you have memory to spare. The VRAM pages below list what fits each GPU.
How much VRAM does a DeepSeek model need?
Q4_K_M files of DeepSeek models on this site range from 0.0 to 954.6 GB. Add the KV cache for your context length and about 0.5 GB for the runtime, which usually means 1.5 to 3 GB on top of the file at an 8K context.
Where do DeepSeek GGUF files come from?
The most common quantizers here are mradermacher, roleplaiapp and Thireus. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.