GGUF models by ukung
43 GGUF models quantized by ukung on Hugging Face. Sizes run from 135M to 14B parameters. The longest trained context is 256K tokens.
About ukung on Hugging Face
ukung has published 43 GGUF models that ggufmodels.com tracks, across families such as Qwen, Nusantara and Llama. The most downloaded are komodo 7b base, Nusantara 4b Indo Chat and gte-Qwen1.5-7B-instruct.
The quants ukung publishes most often are Q8_0 (37), Q4_K_M (34), Q4_0 (33) and Q3_K_L (33). Their Q4_K_M files range from 0.2 to 8.6 GB. Parameter counts run from 135M to 14B, and the longest trained context window is 256K tokens.
Frequently asked questions
What GGUF models has ukung published?
43 models are listed here, mainly from the Qwen, Nusantara and Llama families. The most downloaded is komodo 7b base.
Which quants does ukung publish?
The most common are Q8_0 (37 models), Q4_K_M (34 models), Q4_0 (33 models) and Q3_K_L (33 models).