GGUF models by LoneStriker
163 GGUF models quantized by LoneStriker on Hugging Face. Sizes run from 1.1B to 236B parameters. The longest trained context is 1024K tokens.
About LoneStriker on Hugging Face
LoneStriker has published 163 GGUF models that ggufmodels.com tracks, across families such as Llama, Mistral and Gemma. The most downloaded are deepseek coder 7b instruct v1.5, DeepSeek-Coder-V2-Lite-Instruct and Mistral 7B Instruct v0.2 Neural Story.
The quants LoneStriker publishes most often are Q4_K_M (159), Q3_K_L (154), Q5_K_M (149) and Q6_K (138). Their Q4_K_M files range from 0.7 to 142.5 GB. Parameter counts run from 1.1B to 236B, and the longest trained context window is 1024K tokens.
Frequently asked questions
What GGUF models has LoneStriker published?
163 models are listed here, mainly from the Llama, Mistral and Gemma families. The most downloaded is deepseek coder 7b instruct v1.5.
Which quants does LoneStriker publish?
The most common are Q4_K_M (159 models), Q3_K_L (154 models), Q5_K_M (149 models) and Q6_K (138 models).