ComfyUI INT8 ConvRot Quants
Models for use with ComfyUI.
Model source
- Kind
- UNet
- File size
- 8.76 GB
- Version
- vINT8
- Updated
- Oct 25, 2026
- Downloads
- 0
- Source
- Hugging Face
Source description
Models for use with ComfyUI.
INT8-ConvRot converted from either FP32 or BF16 weights using the conversion script from https://github.com/BobJohnson24/ComfyUI-INT8-Fast
I'm just converting image models for now. Video models, like LTX2.3, seem to have issues with blurry outputs when using INT8-ConvRot.
Leave your reqeusts in the Community tab - I'm happy to quant any model for you and upload it here!
Sources
1 source
Sources
1 sourceVerified Oct 25
Model artifacts
1 artifact
Model artifacts
1 artifactdiffusion_models/flux2_klein_9b_base_int8_convrot.safetensors
safetensors · INT8 (unverified) · 8.76 GB · SHA-256 1675b001e4ee…a356 · Hugging Face
Download fileSource excerpts
3 excerpts
Source excerpts
3 excerptsModels for use with ComfyUI. INT8-ConvRot converted from either FP32 or BF16 weights using the conversion script from https://github.com/BobJohnson24/ComfyUI-INT8-Fast I'm just converting image models for now. Video models, like LTX2.3, seem to have issues with blurry outputs when using INT8-ConvRot. Leave your reqeusts in the Community tab - I'm happy to quant any model for you and upload it here!
Source context: 0 downloads · 5 likes · Repo Se0ulSeeker/ComfyUI-INT8-ConvRot-Quants
Source context: 0 downloads · 4 likes · Repo Se0ulSeeker/ComfyUI-INT8-ConvRot-Quants
