Resourcesrepository2026-06-20Safety reviewed
200+ Image and Video Models, Running on Your Own Machine
The self-hosted alternative to the big AI-video platforms... 200+ models, your hardware, your rules.
- 20.6K stars
- JavaScript
- MIT
- 200+AI models
- 60+video models
- 9lip-sync models
Open Generative AI bundles image, video, and lip-sync generation into one self-hosted studio... 50+ text-to-image models, 60+ image-to-video, nine lip-sync engines, with local inference accelerated on Apple Silicon.
Because it runs on your machine, source material never leaves it... and you are never waiting on someone else's queue, quota, or pricing page. The whole studio is yours to run.
What it does
- Generate images with 50+ text-to-image models
- Transform images with 55+ image-to-image editing models
- Create video from text across 40+ video models
- Animate still frames into video with 60+ models
- Sync audio to portraits or video with 9 lip-sync models
- Upload up to 14 reference images for compatible models
- Build multi-step AI pipelines in a visual workflow studio
- Pro cinema controls: lens, aperture, focal length
- Local model inference on Mac, Windows, Linux with sd.cpp
- Run remote video models via a Wan2GP server
- Browse and reuse previously uploaded images instantly
- Download generated images and video at full resolution
- API keys stored securely in browser localStorage
- Desktop apps for macOS, Windows, and Linux
- Self-host entirely... no subscription, fully customizable
- Community templates and workflows out of the box
Text-to-image models50+ total
- Flux Dev
- Nano Banana 2
- Seedream 5.0
- Ideogram v3
- Midjourney v7
- GPT-4o
- SDXL
- Kling v3
- Veo 3
- Wan 2.6
- Z-Image Turbo
- Z-Image Base
- Dreamshaper 8
- Realistic Vision v5.1
- Anything v5
- SDXL Base 1.0
- MiniMax Image 01
- Flux 1 Dev
- Qwen Image
Image-to-image models55+ total
- Nano Banana 2 Edit
- Flux Kontext Dev
- GPT-4o Edit
- Seededit v3
- Upscaler
- Background Remover
- Kontext
- Seedream 5.0 Edit
- Flux 2 Pro Edit
- Wan 2.6 Image Edit
- Qwen Image Edit Plus
Text-to-video models40+ total
- Kling v3
- Sora 2
- Veo 3
- Wan 2.6
- Seedance 2.0
- Seedance 2.0 Extend
- Hailuo 2.3
- Runway Gen-3
- Grok Imagine T2V
- MiniMax Hailuo 02