Modifying the current pipeline to perform video to video synthesis

[Vasu-Eranki]

Was trying to work on modifying the current pipeline to see if its possible to do video-to-video generation from the current text-to-video pipeline.

Is there a way to open a branch, I'd love to get your opinion on the pipeline to see if I'm going in the right direction, or would you recommend I fork the repo and make the edits to the forked repository.

[ved-genmo]

It is definitely possible to do video-to-video generation. If you make a fork, I'd be happy to take a look :)
This is good prior work in the space: https://github.com/logtd/ComfyUI-MochiEdit