GGUF models by leafspark
36 GGUF models quantized by leafspark on Hugging Face. Sizes run from 121M to 218B parameters. The longest trained context is 1000K tokens.
About leafspark on Hugging Face
leafspark has published 36 GGUF models that ggufmodels.com tracks, across families such as Yi, Llama and Wikichat. The most downloaded are Llama 3.2 11B Vision Instruct, Llama-3-8B-Instruct-Gradient-4194k and Reflection Llama 3.1 70B.
The quants leafspark publishes most often are Q4_K_M (11), Q8_0 (9), Q2_K (9) and Q6_K (7). Their Q4_K_M files range from 4.1 to 50.7 GB. Parameter counts run from 121M to 218B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has leafspark published?
36 models are listed here, mainly from the Yi, Llama and Wikichat families. The most downloaded is Llama 3.2 11B Vision Instruct.
Which quants does leafspark publish?
The most common are Q4_K_M (11 models), Q8_0 (9 models), Q2_K (9 models) and Q6_K (7 models).