GGUF models by Nexesenex
44 GGUF models quantized by Nexesenex on Hugging Face. Sizes run from 1.24B to 94.6B parameters. The longest trained context is 195K tokens.
About Nexesenex on Hugging Face
Nexesenex has published 44 GGUF models that ggufmodels.com tracks, across families such as Llama, Yi and Mistral. The most downloaded are brucethemoose Yi 34B 200K DARE megamerge v8 iMat, Google_Gemma_2_9b_it and brucethemoose Yi 34B 200K DARE merge v7 iMat.
The quants Nexesenex publishes most often are Q8_0 (32), Q4_K_S (19), Q4_K_M (18) and Q3_K_M (17). Their Q4_K_M files range from 4.1 to 20.7 GB. Parameter counts run from 1.24B to 94.6B, and the longest trained context window is 195K tokens.
Frequently asked questions
What GGUF models has Nexesenex published?
44 models are listed here, mainly from the Llama, Yi and Mistral families. The most downloaded is brucethemoose Yi 34B 200K DARE megamerge v8 iMat.
Which quants does Nexesenex publish?
The most common are Q8_0 (32 models), Q4_K_S (19 models), Q4_K_M (18 models) and Q3_K_M (17 models).