Captured photo
mochi-1
140

Percs

High quality
9:16 vertical

About

Mochi 1 is an open-source text-to-video model designed to generate videos with detailed motion and strong adherence to text descriptions. It produces high-fidelity video output suitable for creative projects, animations, and visual content creation. The model emphasizes accurate prompt interpretation and smooth motion rendering.

Settings

Frame Count-  Number of frames to generate
Inference Steps-  Number of inference steps
Guidance Scale-  The guidance scale for the model
FPS-  Frames per second
Seed-  Random seed