GGUF models by newsletter
56 GGUF models quantized by newsletter on Hugging Face. Sizes run from 596M to 46.7B parameters. The longest trained context is 1000K tokens.
About newsletter on Hugging Face
newsletter has published 56 GGUF models that ggufmodels.com tracks, across families such as Mistral, Gemma and Phi. The most downloaded are Phi 3 Medium 128k Instruct, Phi 3 mini 128k instruct Q6 K and CapybaraHermes 2.5 Mistral 7B Q6 K.
The quants newsletter publishes most often are Q6_K (42), Q8_0 (13) and Q4_K_M (1). Their Q4_K_M files range from 21.5 to 21.5 GB. Parameter counts run from 596M to 46.7B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has newsletter published?
56 models are listed here, mainly from the Mistral, Gemma and Phi families. The most downloaded is Phi 3 Medium 128k Instruct.
Which quants does newsletter publish?
The most common are Q6_K (42 models), Q8_0 (13 models) and Q4_K_M (1 models).