Mellum GGUF models
91 GGUF models in the Mellum family, quantized by 35 uploaders. Sizes run from 833M to 12.6B parameters. The longest trained context is 128K tokens.
About the Mellum GGUF models
Mellum is represented on ggufmodels.com by 91 GGUF models, quantized by 35 uploaders on Hugging Face. The most downloaded are Mellum2.1-12B-A2.5B-Thinking, Mellum2 12B A2.5B Instruct and Mellum2 12B A2.5B Instruct. Parameter counts run from 833M to 12.6B, and the longest trained context window is 128K tokens.
A Q4_K_M file of a Mellum model is between 0.5 to 9.0 GB, which is the quant most people start with. Most files come from mradermacher (24), JetBrains (17) and kingjones777 (6). The quants you will meet most often are Q4_K_M (57), Q8_0 (50), Q6_K (47) and Q5_K_M (42), where the number is how many models offer that quant.
Frequently asked questions
Which Mellum GGUF should I download?
The most downloaded Mellum GGUF here is Mellum2.1-12B-A2.5B-Thinking. 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 Mellum model need?
Q4_K_M files of Mellum models on this site range from 0.5 to 9.0 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 Mellum GGUF files come from?
The most common quantizers here are mradermacher, JetBrains and kingjones777. They convert the original model and publish quantized files on Hugging Face. ggufmodels.com does not host files; every download button opens Hugging Face.