GGUF models by Goekdeniz-Guelmez
54 GGUF models quantized by Goekdeniz-Guelmez on Hugging Face. Sizes run from 596M to 14.8B parameters. The longest trained context is 128K tokens.
About Goekdeniz-Guelmez on Hugging Face
Goekdeniz-Guelmez has published 54 GGUF models that ggufmodels.com tracks, across families such as Qwen, Mistral and Stage. The most downloaded are Qwen2 1.5B Instruct, Josiefied Qwen2.5 1.5B Instruct abliterated v3 and Josiefied Qwen2.5 14B Instruct Abliterated v4.
The quants Goekdeniz-Guelmez publishes most often are Q4_K_M (45), Q4_K_S (41), Q6_K (40) and Q5_K_M (39). Their Q4_K_M files range from 0.5 to 9.0 GB. Parameter counts run from 596M to 14.8B, and the longest trained context window is 128K tokens.
Frequently asked questions
What GGUF models has Goekdeniz-Guelmez published?
54 models are listed here, mainly from the Qwen, Mistral and Stage families. The most downloaded is Qwen2 1.5B Instruct.
Which quants does Goekdeniz-Guelmez publish?
The most common are Q4_K_M (45 models), Q4_K_S (41 models), Q6_K (40 models) and Q5_K_M (39 models).