GGUF models by Ransss
119 GGUF models quantized by Ransss on Hugging Face. Sizes run from 7.24B to 30.7B parameters. The longest trained context is 1000K tokens.
About Ransss on Hugging Face
Ransss has published 119 GGUF models that ggufmodels.com tracks, across families such as Llama, Mistral and Forgotten. The most downloaded are DarkIdol Llama 3.1 8B Instruct 1.0 Uncensored Q8 0, Forgotten Safeword 12B v4.0 Q8 0 and L3-Umbral-Mind-RP-v0.3-8B.
The quants Ransss publishes most often are Q8_0 (95) and Q6_K (24). Parameter counts run from 7.24B to 30.7B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has Ransss published?
119 models are listed here, mainly from the Llama, Mistral and Forgotten families. The most downloaded is DarkIdol Llama 3.1 8B Instruct 1.0 Uncensored Q8 0.
Which quants does Ransss publish?
The most common are Q8_0 (95 models) and Q6_K (24 models).