This project is a comfyui node adaption of Consistency Decoder (https://github.com/openai/consistencydecoder).
- 20231107 Support multiple images but it would exceed 24GB vram usage. Please don't use with batch size more than 2.
If you are using venv, please activate venv first.
pip install git+https://github.com/openai/consistencydecoder.git- goto 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/clip- 24GB vram at peak
It only works with SD1.5 latent mapping
Original Image:
Consistency Decoder:
