This project is a comfyui node adaption of Consistency Decoder
Git repo:
https://github.com/lrzjason/ConsistencyDecoderNode
Installation
Please activate venv first if neccessary.
pip install git+https://github.com/openai/consistencydecoder.gitgoto ComfyUI\custom_nodes
git clone https://github.com/lrzjason/ConsistencyDecoderNodePlease download the model from following link
https://openaipublic.azureedge.net/diff-vae/c9cebd3132dd9c42936d803e33424145a748843c8f716c0814838bdc8a2fe7cb/decoder.ptplace the model to
C:\\Users\\[user]/.cache/clipHardware Requirement
24GB vram at peak
Description
Tool
FAQ
Comments (4)
Its not clear what is it for? Does it improve what aspect of image generation? It should be only used in img2img to increase consistency?
It is a vae replacement developed by openai. I just make the node adaption for it.
What's the difference between yours, and this other one?
https://github.com/glifxyz/ComfyUI-GlifNodes
I don't know. Might be no difference. I just implemented the official implementation in comfyui.