Mochi 1 Preview - AI Vision Models Tool

Overview

Mochi 1 Preview is an open text-to-video generation model from Genmo. It uses a 10 billion parameter diffusion model with an Asymmetric Diffusion Transformer to produce high-fidelity videos from text prompts and is released under the Apache 2.0 license.

Key Features

• 10 billion parameter diffusion model for text-to-video generation.
• Asymmetric Diffusion Transformer architecture for high-fidelity video synthesis.
• Open-source release under the Apache 2.0 license.
• Hosted on Hugging Face for model access and downloads.

Ideal Use Cases

• Generate short videos from text prompts for creative prototyping.
• Experiment with diffusion-based video synthesis research.
• Customize base model for experimental video pipelines.
• Distribute and modify model artifacts under Apache 2.0.

Getting Started

• Open the model page on Hugging Face: https://huggingface.co/genmo/mochi-1-preview.
• Read the model card and Apache 2.0 license details.
• Follow provided examples to load the model into your environment.
• Provide text prompts and run inference to generate videos.

Pricing

Pricing not disclosed.

Limitations

• Preview release — features, performance, or API may change.

Key Information

• Category: Vision Models
• Type: AI Vision Models Tool