Example workflow ic-lora not working properly #267
Description
I use ComfyUI portable in Windows 10, everything updated to the latest version. 16Gb VRAM & 64Gb RAM.
When I try to use the example ic-lora workflow (with all nodes at theyre default settings, default models, default inputs etc.), it will always give strange results.
The final video output becomes a combination of the original video input, combined with the resulting depth, pose or canny video. Never a proper result with the robot walking.
Trying to find out what is going wrong is beyond my competence.
It would be great if you could take a look at it and provide an updated version of this workflow that will run "out of the box".
Many thanks in advance!