You can now run MiniMax-H3 in Unsloth!
#5
by danielhanchen - opened
We're optimizing it even further! Also work sin stablediffusion.cpp
GitHub: https://github.com/unslothai/unsloth
Why are there two mouses in the animation?
Unsloth's quants also work in ComfyUI, the GGUF loader nodes load them fine :)
I can't load the clip model download from this repo.
ValueError: This gguf file is incompatible with llama.cpp!
Consider using safetensors or a compatible gguf file
Unsloth's quants also work in ComfyUI, the GGUF loader nodes load them fine :)
I can't load the clip model download from this repo.
ValueError: This gguf file is incompatible with llama.cpp! Consider using safetensors or a compatible gguf file
Don't use the clip model in this repo. Follow the "GGUF loader nodes" Link.
Clone it into your custom nodes folder. And download the" Qwen3-VL-32B-Instruct-MiniMax-H3 pruned GGUFs" model listed in the readme.