GGUF models by mudler
92 GGUF models quantized by mudler on Hugging Face. Sizes run from 30M to 397B parameters. The longest trained context is 1024K tokens.
About mudler on Hugging Face
mudler has published 92 GGUF models that ggufmodels.com tracks, across families such as Qwen, Rfdetr and Minerva. The most downloaded are locate anything.cpp, Laguna XS 2.1 APEX and KAT Coder V2.5 Dev APEX.
The quants mudler publishes most often are F16 (33), Q8_0 (20), Q4_K (16) and F32 (15). Their Q4_K_M files range from 1.8 to 8.6 GB. Parameter counts run from 30M to 397B, and the longest trained context window is 1024K tokens.
Frequently asked questions
What GGUF models has mudler published?
92 models are listed here, mainly from the Qwen, Rfdetr and Minerva families. The most downloaded is locate anything.cpp.
Which quants does mudler publish?
The most common are F16 (33 models), Q8_0 (20 models), Q4_K (16 models) and F32 (15 models).