tools
Mochi 1
Mochi 1 is an open-source text-to-video model that turns written prompts into short videos.
Mochi 1 generates short videos from natural-language prompts. Genmo provides a hosted playground, open weights, a GitHub repository, and ComfyUI support for local use.
The model is intended for researchers, developers, artists, educators, advertisers, and other creators. The preview produces 480p video at 30 frames per second for up to 5.4 seconds, may produce warping during extreme motion, and is less suited to animated content. Hosted use consumes credits.
Features
- Generates video from text prompts
- Produces 480p video in the initial preview
- Generates video at 30 frames per second
- Supports clips up to 5.4 seconds
- Provides open weights and architecture
- Licensed under Apache 2.0
- Runs locally through GitHub code and ComfyUI
- Includes prompt moderation in the hosted playground
Use cases
- Create short concept videos from written descriptions
- Prototype shots for films, ads, and educational media
- Research open video-generation architectures
- Generate synthetic video data for robotics experiments
- Customize the model for specific visual styles
- Test text-to-video workflows in a hosted playground