GGUF models by legraphista
138 GGUF models quantized by legraphista on Hugging Face. Sizes run from 494M to 236B parameters. The longest trained context is 1024K tokens.
About legraphista on Hugging Face
legraphista has published 138 GGUF models that ggufmodels.com tracks, across families such as Qwen, Llama and Mistral. The most downloaded are Meta-Llama-3-70B-Instruct-abliterated-v3.5, DeepSeek-V2-Lite and openchat-3.6-8b-20240522.
The quants legraphista publishes most often are Q8_0 (138), Q6_K (138), Q5_K_S (138) and Q5_K (138). Their Q4_K_M files range from 4.1 to 21.5 GB. Parameter counts run from 494M to 236B, and the longest trained context window is 1024K tokens.
Frequently asked questions
What GGUF models has legraphista published?
138 models are listed here, mainly from the Qwen, Llama and Mistral families. The most downloaded is Meta-Llama-3-70B-Instruct-abliterated-v3.5.
Which quants does legraphista publish?
The most common are Q8_0 (138 models), Q6_K (138 models), Q5_K_S (138 models) and Q5_K (138 models).