Imagine AI that can simulate anything—possible or impossible.

Our video generation models act as world simulators, driving breakthroughs in embodied AI by enabling infinite explorations in synthetic realities. Video is the ultimate medium for human-AI interaction, seamlessly integrating text, audio, images, and 3D into one unified experience.

Our team includes the original creators of DDPM, DreamFusion, and Emu Video.

Mochi 1, our first public open-source release, is licensed under Apache 2.0 for both individual and commercial use.

Our principles:

1. 1State-of-the-Art Performance: Reality sets a very high standard for video generation. We are going to close the gap.
2. 2Open Models: Open ecosystems win over closed. Open-source is the best long-term business model for video generation.
3. 3Community-first: By releasing Mochi 1 under the Apache 2.0 license, the community can build on it by sharing their fine-tunes.