Pruned layers after 49 for H3.
You need only the model for T2V and model + mmproj for I2V. place both the model and the mmproj file in text_encoder folder.
Can be used in ComfyUI with my fork of City96's ComfyUI-GGUF. https://github.com/Nif00/ComfyUI-GGUF
Have the files? Submit a torrent.
Submit a torrentHave the complete files and permission to share them? Replace the example folder path. The script checks the files, makes a torrent and submits its magnet and checksums. Keep your torrent client seeding. Get your community key from your settings, and read the script before running it.
curl -fsSL https://pirateface.co/package.sh | bash -s -- --repo nif0/Qwen3-VL-32B-Instruct-ultra-uncensored-heretic-Minimax-H3-GGUF ./model-foldermradermacher/Qwen3-VL-32B-Instruct-ultra-uncensored-heretic-i1-GGUF