About this project

Open Generative AI is a free, open-source creative studio for AI image, video, audio, and lip sync generation. Powered by Muapi.ai, it provides access to 400+ models (including Flux, Midjourney, Kling, Sora, Veo, and Seedream) across 14 distinct studios. The application emphasizes unrestricted creative freedom, operating without content filters, prompt rejections, or subscription fees. Key studios and features include: - Image Studio: Supports text-to-image (50+ models) and image-to-image (55+ models) generation, with multi-image input allowing up to 14 reference images for compatible models. - Video Studio: Offers text-to-video (40+ models) and image-to-video (60+ models) generation. - Lip Sync Studio: Animates portraits or syncs lips to existing videos using audio, featuring 9 dedicated models. - Additional studios: Audio Studio, AI Clipping, Vibe Motion, Body Swap, Cinema, Marketing, Workflow, and Agent studios. The platform supports local model inference via two engines: sd.cpp (bundled, for SD 1.5, SDXL, and Z-Image models on Mac/Win/Linux) and Wan2GP (a remote Gradio server for Flux and video models requiring CUDA/ROCm GPUs). It offers a desktop app for macOS, Windows, and Linux, alongside a hosted web version. The UI features a dark glassmorphism design, generation history, and smart controls that adapt to selected models.