Add Mochi Video Model

[ghunkins]

Model/Pipeline/Scheduler description

A new state-of-the-art open-source video model was released today under the Apache-2.0 license: Mochi.

Would be fantastic to get support for this model via a pipeline in Diffusers.

Blog | Hugging Face | Github

A state of the art video generation model by Genmo.

https://github.com/user-attachments/assets/4d268d02-906d-4cb0-87cc-f467f1497108

Open source status

• [X] The model implementation is available.
• [X] The model weights are available (Only relevant if addition is not a scheduler).

Provide useful links for the implementation

Links: Github | Inference Script

Authors: @ved-genmo @akbar-genmo @paras-genmo

[a-r-r-o-w]

Already on it :)

[HashingTag]

Kijai has released optimizations for low VRAM here: https://github.com/kijai/ComfyUI-MochiWrapper

• some extra information here: https://reddit.com/r/StableDiffusion/comments/1g9n9kf/comment/lt8spp5