Innovate Futures @ Benji
patreon
[Workflow] Tora CogVideoX img2vid Unsampling SEGs Detailer 20241025
🕑 Added 2024-10-25 12:55:43 +0000 UTC
Related Video: https://youtu.be/fjq75fLX9SQ
This is an update workflow for img2vid using Tora CogVideoX.
In this workflow, I added AnimateDiff Unsampling and Sampling Part With SEGs Detailer to fix morphing hands or faces.
In ComfyUI , you will need ComfyUI-CogVideoXWrapper
https://github.com/kijai/ComfyUI-CogVideoXWrapper
Download Tora for Cog. https://huggingface.co/Kijai/CogVideoX-5b-Tora/tree/main
[Update] Some models needed in this workflow
Hotshot XL for AnimateDiff Model : https://huggingface.co/hotshotco/Hotshot-XL/blob/main/hsxl_temporal_layers.f16.safetensors
Controlnet-union-sdxl-1.0
https://huggingface.co/xinsir/controlnet-union-sdxl-1.0
Tutorial for SDXL union controlnet promax : https://www.youtube.com/watch?v=R14aPPjsiRM
Any question, feel free to ask in the comment section.
Thanks.
Comments
Benjamin Law
hi, I have updated the post, I list models that need for the AnimateDiff, and ControlNet for SDXL. thanks
MarcAndreMueller
somehow after updating im still missing a couple of nodes in your workflow. the ones that say Get CHQ model, get AD model and a bunch of others. any advice on how to fix it?
Clayfacer
Awesome. Thank you
