GGUF models by hus960
47 GGUF models quantized by hus960 on Hugging Face. Sizes run from 2.78B to 58.3B parameters. The longest trained context is 1024K tokens.
About hus960 on Hugging Face
hus960 has published 47 GGUF models that ggufmodels.com tracks, across families such as Llama, Mistral and Qwen. The most downloaded are dolphin 2.8 mistral 7b v02 Q4 K M, Llama-3-13B-Instruct and Einstein-v6.1-Llama3-8B.
The quants hus960 publishes most often are Q4_K_M (44), Q8_0 (1), Q2_K (1) and BF16 (1). Their Q4_K_M files range from 4.1 to 28.4 GB. Parameter counts run from 2.78B to 58.3B, and the longest trained context window is 1024K tokens.
Frequently asked questions
What GGUF models has hus960 published?
47 models are listed here, mainly from the Llama, Mistral and Qwen families. The most downloaded is dolphin 2.8 mistral 7b v02 Q4 K M.
Which quants does hus960 publish?
The most common are Q4_K_M (44 models), Q8_0 (1 models), Q2_K (1 models) and BF16 (1 models).