Enfugue is an open-source web application for creating studio-grade images and video with generative AI, built around Stable Diffusion. It packages a full creative workstation into a self-hosted app you run on your own machine, and it has collected about 754 stars on GitHub. The project was last updated in October 2024.
Where many diffusion tools are either bare command line scripts or hosted services, Enfugue aims for a polished, self-contained studio experience. It runs as a local server with a browser interface, ships as a portable executable, a PyPI package and a Docker image, and works across Windows, macOS and Linux. It supports NVIDIA GPUs as well as Apple Silicon through MPS, so creators can generate locally without sending work to a third-party cloud. The interface is designed for iterative, professional image work rather than one-off prompt experiments.
Because everything runs locally, generation speed depends on your own GPU, and your library of models and outputs stays under your control. The portable build means the whole studio can live on an external drive and move between machines without installation steps, accounts or activation, which is unusual among diffusion tools of this completeness.
Enfugue is free and open source under the GPL-3.0 license. There is no subscription or hosted tier; you download it, run it on your own hardware, and only pay for the compute you already own.
Enfugue is for digital artists, designers and AI image enthusiasts who want a proper studio tool for Stable Diffusion without relying on paid cloud generators. It also fits privacy-conscious creators who prefer that prompts and outputs never leave their own computer.