GGUF models by Abiray
108 GGUF models quantized by Abiray on Hugging Face. Sizes run from 134M to 197B parameters. The longest trained context is 1024K tokens.
About Abiray on Hugging Face
Abiray has published 108 GGUF models that ggufmodels.com tracks, across families such as Qwen, Gemma and Minimax. The most downloaded are MiniMax H3 Pruned, MiniMax H3 and DeepSeek V4.1 Flash 1017 Uncensored GSQ RCO IQ2 XXS.
The quants Abiray publishes most often are Q4_K_M (98), Q5_K_M (86), Q6_K (85) and Q8_0 (80). Their Q4_K_M files range from 0.3 to 75.2 GB. Parameter counts run from 134M to 197B, and the longest trained context window is 1024K tokens.
Frequently asked questions
What GGUF models has Abiray published?
108 models are listed here, mainly from the Qwen, Gemma and Minimax families. The most downloaded is MiniMax H3 Pruned.
Which quants does Abiray publish?
The most common are Q4_K_M (98 models), Q5_K_M (86 models), Q6_K (85 models) and Q8_0 (80 models).