Run Mochi in ComfyUI

Announcement
1

ComfyUI now has optimized support for Genmo’s latest video generation model, Mochi! Now it runs natively in a consumer GPU! To run the Mochi model right away with a standard workflow, try the following steps.

  1. Update to the latest version of ComfyUI
  2. Download Mochi weights (the diffusion models) into models/diffusion_model folder
  3. Make sure a text encoder is in your models/clip folder
  4. Download VAE to: ComfyUI/models/vae
  5. Grab the workflow from and dive into creation!

We’ve also updated our Example Workflows page with text encoder and VAE setups for Mochi. Full blog post: https://blog.comfy.org/mochi/ Huge shoutout to @Kijai who did the initial implementation in his node, what a legend!!

1 Like