GGUF models by Ak1104
35 GGUF models quantized by Ak1104 on Hugging Face. Sizes run from 6.74B to 22.2B parameters. The longest trained context is 256K tokens.
About Ak1104 on Hugging Face
Ak1104 has published 35 GGUF models that ggufmodels.com tracks, across families such as Llama, Cde and Withoutchapter. The most downloaded are codellama 7b rust, QA_withoutChapter and codellama_cde.
The quants Ak1104 publishes most often are Q4_K_M (31), F16 (24), Q8_0 (2) and Q5_K_M (1). Their Q4_K_M files range from 4.1 to 13.3 GB. Parameter counts run from 6.74B to 22.2B, and the longest trained context window is 256K tokens.
Frequently asked questions
What GGUF models has Ak1104 published?
35 models are listed here, mainly from the Llama, Cde and Withoutchapter families. The most downloaded is codellama 7b rust.
Which quants does Ak1104 publish?
The most common are Q4_K_M (31 models), F16 (24 models), Q8_0 (2 models) and Q5_K_M (1 models).