Repolode
Submit
Anil-matcha

Anil-matcha/Open-Generative-AI

Self-hostable Next.js and Electron studio for generating AI images, videos and lip-sync clips through the Muapi.ai API, with no content filters.

Open-source alternative to:MidjourneyRunwayKling AI

View on GitHubWebsite
Stars
32,806
Forks
6,232
Open issues
46
Last commit
yesterday

What is it?

Open Generative AI is a studio interface for generating images, videos, and lip-synced talking clips with AI models. It is built with Next.js and packaged as a desktop app via Electron, and it sends generation requests to the Muapi.ai API, which provides access to a large catalog of hosted image and video models such as Flux, Kling, Sora, Veo, and Wan.

Beyond cloud generation, the desktop app can run local image models through a bundled sd.cpp engine (SD 1.5, SDXL, Z-Image) and can route video models to a user-installed Wan2GP instance on a local folder or remote MCP server. The interface includes separate studios for image, video, audio, lip sync, cinema-style camera controls, multi-step workflows, and agent-driven generation, along with upload history and generation history stored in the browser.

The project is MIT licensed and designed to be forked and modified. Using the hosted API requires a Muapi.ai access key, while the local engines work without one.

Key features

  • 01Image and video generation from text prompts and reference images across many hosted models.
  • 02Lip Sync Studio that turns portrait images or videos into talking clips using uploaded audio.
  • 03Local image generation through a bundled sd.cpp engine on Mac, Windows, and Linux.
  • 04Video generation via a self-installed Wan2GP instance, locally or on a remote machine over MCP.
  • 05Multi-image input of up to 14 reference images for compatible edit models.
  • 06Node-based Workflow Studio for chaining image, video, and audio models into pipelines.

Use cases

  • Producing marketing or social media image and video variations without a paid subscription.
  • Making talking-head videos from a portrait and an audio track.
  • Running Stable Diffusion-family image models locally on a Mac.
  • Self-hosting a generative media studio that can be customized for a team or product.

Quick start

  1. 1

    Download the desktop installer for your platform from the releases page

  2. 2

    On macOS, clear the quarantine flag after moving the app to Applications

    xattr -cr "/Applications/Open Generative AI.app"

Things to know

  • The app is not notarized by Apple, so macOS Gatekeeper blocks it on first launch.
  • The Windows installer is not code-signed, so SmartScreen shows a warning.
  • Generation runs through the Muapi.ai API, so cloud features depend on that service.
  • Local model inference needs substantial memory, and video models need an NVIDIA or AMD GPU via Wan2GP.

In collections

AI video generationSelf-hosted appsAI image generators & Stable Diffusion tools

Topics

#ai-image-generation#ai-video-generation#lip-sync#text-to-video#text-to-image#nextjs#electron#self-hosted#ai-art-generator#fal-ai-alternative

FAQ

What is Open-Generative-AI?
Self-hostable Next.js and Electron studio for generating AI images, videos and lip-sync clips through the Muapi.ai API, with no content filters. Open Generative AI is a studio interface for generating images, videos, and lip-synced talking clips with AI models. It is built with Next.js and packaged as a desktop app via Electron, and it sends generation requests to the Muapi.ai API, which provides access to a large catalog of hosted image and video models such as Flux, Kling, Sora, Veo, and Wan.
Is Open-Generative-AI free to use? What license does it have?
Open-Generative-AI is open source under the MIT license. Check the license terms before using it in commercial products.
Is Open-Generative-AI still maintained?
Actively maintained. The last commit to Open-Generative-AI was on Nov 6, 2026.
What can I use Open-Generative-AI for?
Producing marketing or social media image and video variations without a paid subscription. Making talking-head videos from a portrait and an audio track. Running Stable Diffusion-family image models locally on a Mac. Self-hosting a generative media studio that can be customized for a team or product.
How do I install Open-Generative-AI?
Download the desktop installer for your platform from the releases page On macOS, clear the quarantine flag after moving the app to Applications: xattr -cr "/Applications/Open Generative AI.app"
Where can I get Open-Generative-AI?
The source code and installation instructions are on GitHub: https://github.com/Anil-matcha/Open-Generative-AI

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects