Open Generative AI is a free, open-source image and video generation studio that puts creative control entirely in your hands. Unlike proprietary platforms that gate models behind subscriptions, enforce content filters, and lock you into a closed ecosystem, Open Generative AI gives you direct access to 200+ state-of-the-art models — including Flux, Midjourney, Kling, Sora, Veo, and Seedance — from a single, self-hostable interface. Whether you’re an artist pushing creative boundaries, a developer building on top of AI media pipelines, or a researcher who needs unrestricted experimentation, Open Generative AI is built for you. It runs on your machine, respects your data, and is MIT licensed so you can fork, extend, and ship it however you like.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/anil-matcha/open-generative-ai/llms.txt
Use this file to discover all available pages before exploring further.
Key Features
200+ Models
Access more than 200 text-to-image, image-to-image, text-to-video, image-to-video, and lip sync models — including Flux, Midjourney v7, Kling v3, Sora 2, Veo 3, Seedance 2.0, and GPT-4o — all from one unified interface.
No Content Filters
No prompt rejections, no guardrails, no nanny filters. Generate exactly what you intend without having your creative work blocked or silently altered by a platform policy.
Self-Hostable
Run the full studio on your own machine or server. Your prompts and generated media never leave your infrastructure unless you choose otherwise. Docker support included.
Desktop App
One-click installers for macOS (Apple Silicon and Intel), Windows, and Linux — no Node.js or terminal required. Download and start generating in minutes.
Local Inference
Run image generation entirely offline with the bundled sd.cpp engine (Metal on Apple Silicon, CUDA/Vulkan/ROCm on Linux/Windows), or connect to a Wan2GP Gradio server for large video models.
Multi-Image Input
Feed up to 14 reference images into compatible models like Nano Banana 2 Edit, Flux Kontext Dev, and GPT-4o Edit. Images are sent in the order you select them for fine-grained control.
Lip Sync Studio
Animate portrait photos or sync lips on existing videos using audio. Choose from 9 dedicated models across two modes: portrait image + audio, and video + audio.
Workflow Studio
Build and run multi-step AI pipelines with a node-based visual editor. Chain image, video, and audio models into automated flows, browse community templates, and run any workflow interactively.
How It Compares
Open Generative AI is designed as a community-driven alternative to closed AI video and image platforms. Here’s how it stacks up:| Feature | Other Platforms | Open Generative AI |
|---|---|---|
| Cost | Subscription-based | Free (open-source) |
| Content filters | Yes — prompts blocked or altered | None |
| Self-hosting | No | Yes |
| Models | Proprietary, limited selection | 200+ open & commercial models |
| Lip sync | No | 9 models, image & video modes |
| Source code | Closed | MIT licensed |
Tech Stack
Open Generative AI is built on a modern, well-supported set of tools:- Next.js 15 — App Router, server components, and a fast development server
- React 19 — All studio UI components
- Tailwind CSS v3 — Utility-first styling with a dark glassmorphism design language
- npm workspaces monorepo — Shared
packages/studiocomponent library consumed by both the self-hosted app and the hosted version on muapi.ai - Muapi.ai API — Unified API gateway for all 200+ AI image and video models
Explore the Studios
Ready to start generating? Head to the Quickstart to get the app running, then explore the individual studios:Quickstart
Clone the repo, run setup, and generate your first image or video in under 5 minutes.
Image Studio
50+ text-to-image models and 55+ image-to-image models with multi-image input support.
Video Studio
40+ text-to-video and 60+ image-to-video models including Kling, Sora, Veo, and Seedance.
Deployment
Deploy via Docker, download the desktop app, or run the Next.js dev server.