GGUF models by jeiku
89 GGUF models quantized by jeiku on Hugging Face. Sizes run from 2.17B to 16B parameters. The longest trained context is 1000K tokens.
About jeiku on Hugging Face
jeiku has published 89 GGUF models that ggufmodels.com tracks, across families such as Qwen, Aura and Rosa. The most downloaded are ToxicNoRobotsRosaHermesBoros 3B, Luna 7B and sausage.
The quants jeiku publishes most often are Q6_K (41), Q4_K (41), Q5_K (39) and F16 (38). Their Q4_K_M files range from 1.7 to 4.9 GB. Parameter counts run from 2.17B to 16B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has jeiku published?
89 models are listed here, mainly from the Qwen, Aura and Rosa families. The most downloaded is ToxicNoRobotsRosaHermesBoros 3B.
Which quants does jeiku publish?
The most common are Q6_K (41 models), Q4_K (41 models), Q5_K (39 models) and F16 (38 models).