GGUF models by DuoNeural
120 GGUF models quantized by DuoNeural on Hugging Face. Sizes run from 17M to 34.7B parameters. The longest trained context is 1024K tokens.
About DuoNeural on Hugging Face
DuoNeural has published 120 GGUF models that ggufmodels.com tracks, across families such as Qwen, Gemma and DeepSeek. The most downloaded are Gemma4 12B IT Abliterated, LFM2.5 8B A1B Hermes Agentic Coder Abliterated v3 and Cyber Ornith 1.5 9B OBLITERATED.
The quants DuoNeural publishes most often are Q4_K_M (80), IQ3_XXS (40), Q8_0 (35) and IQ2_M (30). Their Q4_K_M files range from 0.0 to 21.2 GB. Parameter counts run from 17M to 34.7B, and the longest trained context window is 1024K tokens.
Frequently asked questions
What GGUF models has DuoNeural published?
120 models are listed here, mainly from the Qwen, Gemma and DeepSeek families. The most downloaded is Gemma4 12B IT Abliterated.
Which quants does DuoNeural publish?
The most common are Q4_K_M (80 models), IQ3_XXS (40 models), Q8_0 (35 models) and IQ2_M (30 models).