GGUF models by backyardai
121 GGUF models quantized by backyardai on Hugging Face. Sizes run from 2.61B to 141B parameters. The longest trained context is 1000K tokens.
About backyardai on Hugging Face
backyardai has published 121 GGUF models that ggufmodels.com tracks, across families such as Llama, Gemma and Chaifighter. The most downloaded are NeuralLemon 7B, WestLake 10.7B v2 and QuartetAnemoi 70B t0.0001.
The quants backyardai publishes most often are Q8_0 (121), Q6_K (121), Q5_K_S (121) and Q5_K_M (121). Their Q4_K_M files range from 1.7 to 85.6 GB. Parameter counts run from 2.61B to 141B, and the longest trained context window is 1000K tokens.
Frequently asked questions
What GGUF models has backyardai published?
121 models are listed here, mainly from the Llama, Gemma and Chaifighter families. The most downloaded is NeuralLemon 7B.
Which quants does backyardai publish?
The most common are Q8_0 (121 models), Q6_K (121 models), Q5_K_S (121 models) and Q5_K_M (121 models).