Mochi 1 by Genmo

Paid

Open-source video generation model for world simulation

4.7

Video GeneratorsContact

Inputs: textOutputs: video

Type

Saas

Company

Genmo

Visit Website

About Mochi 1 by Genmo

Genmo is a research lab dedicated to building frontier models for video generation, aiming to unlock the right brain of artificial general intelligence. Mochi 1 is their first public open-source release, licensed under Apache 2.0 for both individual and commercial use. The model acts as a world simulator, enabling breakthroughs in embodied AI by generating videos from text prompts. The platform offers a free tier with 50 monthly credits and a watermark, as well as paid tiers (Lite and Pro) with higher credit limits, no watermark, commercial usage rights, and priority queue access. Mochi 1 is built on foundational research contributions from the team, including Denoising Diffusion Probabilistic Models and DreamFusion.

Key Features

Open-source under Apache 2.0 license for individual and commercial use

Free tier with 50 monthly credits (includes Genmo watermark)

Lite tier: 1,200 monthly credits, no watermark, commercial usage, standard priority

Pro tier: 5,000 monthly credits, no watermark, highest priority queue, early model access

Mochi video generation costs 100 credits per video; replay video costs 10 credits

Built on state-of-the-art diffusion models from the team's foundational research

Pros & Cons

Pros

Open-source (Apache 2.0) allows flexibility for developers and businesses
Free tier available to start generating videos at no cost
Commercial usage rights included in Lite and Pro plans
Team includes creators of foundational AI research (e.g., diffusion models)
Credit system provides predictable usage-based pricing

Cons

Free tier has watermarked videos and only 50 credits per month
Each Mochi video generation consumes 100 credits, limiting free usage
Requires sign-in via Google or Discord to access the playground
No explicit API mentioned; seems limited to web platform for now

Best For

Generating videos from text prompts for creative projectsEnabling explorations in synthetic realities for AI researchSimulating environments for embodied AI developmentProducing video content for commercial use (on paid tiers)