GGUF models by Agnuxo
93 GGUF models quantized by Agnuxo on Hugging Face. Sizes run from 4M to 12.2B parameters. The longest trained context is 1000K tokens.
About Agnuxo on Hugging Face
Agnuxo has published 93 GGUF models that ggufmodels.com tracks, across families such as Qwen, Tinytron and Llama. The most downloaded are CAJAL 4B Merged 16bit, Mamba Codestral 7B v0.1 instruct python coding assistant GGUF 4bit and Phi 3.5 mini instruct python coding assistant GGUF 16bit.
The quants Agnuxo publishes most often are Q8_0 (26), F16 (25), Q4_K_M (23) and BF16 (19). Their Q4_K_M files range from 0.0 to 5.6 GB. Parameter counts run from 4M to 12.2B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has Agnuxo published?
93 models are listed here, mainly from the Qwen, Tinytron and Llama families. The most downloaded is CAJAL 4B Merged 16bit.
Which quants does Agnuxo publish?
The most common are Q8_0 (26 models), F16 (25 models), Q4_K_M (23 models) and BF16 (19 models).