Mochi 1: A new SOTA in Open-Source Video Generation Models

⁨16⁩ ⁨likes⁩

Submitted ⁨⁨4⁩ ⁨weeks⁩ ago⁩ by ⁨Even_Adder@lemmy.dbzer0.com⁩ to ⁨stable_diffusion@lemmy.dbzer0.com⁩

https://huggingface.co/genmo/mochi-1-preview/resolve/main/assets/grid.gif

Blog: www.genmo.ai/blog

Code: github.com/genmoai/models

Weights: huggingface.co/genmo/mochi-1-preview

Comments

Sort:hotnew top

just_an_average_joe@lemmy.dbzer0.com ⁨4⁩ ⁨weeks⁩ ago
For anyone thinking of running this:

“The model requires at least 4 H100 GPUs to run. We welcome contributions from the community to reduce this requirement.”

Taken from their huggingface card

source
BrianTheeBiscuiteer@lemmy.world ⁨4⁩ ⁨weeks⁩ ago
Will Smith eating spaghetti is the only true test of an AI video creator.

Also, a bit impressed the girl at the top isn’t insanely gorgeous like most AI generations. Making mundane or ugly things is usually a struggle for most models that are built on high quality, studio imagery.

source
- paraphrand@lemmy.world ⁨4⁩ ⁨weeks⁩ ago
  Agreed on both points.
  
  The traditionally attractive factor is another tell with AI generated imagery. One of many that is fading as new models are built.
  
  source