Skip to content

Mochi in ComfyUI

Open video generation inside ComfyUI

Visit official site

What Mochi is

The model is described as an open-source video generation model from Genmo, optimised in ComfyUI to run on consumer GPUs with high-fidelity motion.

What you can do with it

  • Generate video with high-fidelity motion
  • Follow prompts closely for the wanted video
  • Use or modify the model under the Apache 2.0 license

Getting started

  • Update ComfyUI to the latest version
  • Download the weights into the diffusion models folder
  • Place a text encoder in the clip folder and the VAE in the vae folder

Pros & cons

✓ What we like

  • It runs on consumer GPUs such as the 4090
  • Several attention backends fit different VRAM sizes

! What to watch out for

  • Weights and a VAE must be placed in ComfyUI folders
  • The workflow is described as running under 24GB of VRAM

FAQ

What license applies?

The material states the Apache 2.0 license.

Which GPU is named?

The material names a consumer GPU such as the 4090.

Last reviewed: 2026-11-15

More AI video generation tools

View all →

How we review