One GitHub repo just replaced Higgsfield AI, Krea, Freepik, and Openart in one install. Anil-matcha’s Open Generative AI (formerly Open-Higgsfield-AI) is an uncensored, self-hosted, MIT-licensed AI studio with 200+ models built in.
— Quick Note on the Name —
— Old name: Anil-matcha/Open-Higgsfield-AI
— New name: Anil-matcha/Open-Generative-AI
— Same repo, same code. Old URL redirects.
— 4 Studios Built In —
— Image Studio: 50+ text-to-image, 55+ image-to-image models (Flux Dev, Nano Banana 2, Seedream 5.0, Ideogram v3, Midjourney v7, SDXL)
— Video Studio: 40+ text-to-video, 60+ image-to-video models (Kling v3, Sora 2, Veo 3, Wan 2.6, Seedance 2.0, Runway Gen-3)
— Lip Sync Studio: 9 models (Infinite Talk, LTX Lipsync, Wan 2.2 Speech-to-Video, Sync, LatentSync)
— Cinema Studio: photoreal shots with pro camera controls (Lens, Focal Length, Aperture)
— Install in 3 Lines —
git clone https://github.com/Anil-matcha/Open-Generative-AI.git
cd Open-Generative-AI
npm install && npm run dev
Open http://localhost:3000, drop your Muapi API key once, generate forever. Prereqs: Node.js 18+ and a Muapi.ai API key.
— Repo + Hosted App —
— https://github.com/Anil-matcha/Open-Generative-AI
— https://dev.muapi.ai/open-generative-ai (no install needed)
— Free 8-Page Install & Use Guide —
Full breakdown of all 4 studios, model category tables, desktop app builds (macOS / Windows / Linux), local inference with Z-Image Turbo, and workflows:
https://drive.google.com/file/d/182cgslIInv9ggpzKADou6EXr7xAHSCck/view?usp=drivesdk
— Recommended Channels —
3Blue1Brown, Two Minute Papers, freeCodeCamp, Matt Wolfe, Matthew Berman, DeepLearning.AI, Krish Naik, Sentdex, Yannic Kilcher, Nate Herk
#OpenSource #AI #GenerativeAI #Flux #Kling #Sora #Veo #Midjourney #StableDiffusion #AIVideo #LipSync #GitHub #SelfHosted #AITools