GGUF models by Undi95
127 GGUF models quantized by Undi95 on Hugging Face. Sizes run from 7.24B to 72.7B parameters. The longest trained context is 1000K tokens.
About Undi95 on Hugging Face
Undi95 has published 127 GGUF models that ggufmodels.com tracks, across families such as Mistral, Mlewd and Remm. The most downloaded are Lumimaid Magnum v4 12B, MLewd v2.4 13B and Mistral 7B claude chat.
The quants Undi95 publishes most often are Q8_0 (106), Q5_K_M (95), Q6_K (88) and Q4_K_S (81). Their Q4_K_M files range from 4.9 to 47.4 GB. Parameter counts run from 7.24B to 72.7B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has Undi95 published?
127 models are listed here, mainly from the Mistral, Mlewd and Remm families. The most downloaded is Lumimaid Magnum v4 12B.
Which quants does Undi95 publish?
The most common are Q8_0 (106 models), Q5_K_M (95 models), Q6_K (88 models) and Q4_K_S (81 models).