GGUF models by actionpace
134 GGUF models quantized by actionpace on Hugging Face. Sizes run from 3.21B to 70.6B parameters. The longest trained context is 128K tokens.
About actionpace on Hugging Face
actionpace has published 134 GGUF models that ggufmodels.com tracks, across families such as Llama, Remm and Kimi. The most downloaded are Apertus 70B 2509, Apertus 70B Instruct 2509 and llama 2 13b chat limarp v2 merged.
The quants actionpace publishes most often are Q5_1 (77), Q5_K_M (51), Q8_0 (26) and Q4_K_M (21). Their Q4_K_M files range from 5.1 to 43.7 GB. Parameter counts run from 3.21B to 70.6B, and the longest trained context window is 128K tokens.
Frequently asked questions
What GGUF models has actionpace published?
134 models are listed here, mainly from the Llama, Remm and Kimi families. The most downloaded is Apertus 70B 2509.
Which quants does actionpace publish?
The most common are Q5_1 (77 models), Q5_K_M (51 models), Q8_0 (26 models) and Q4_K_M (21 models).