OpenDream (varunshenoy) is an open-source web interface for Stable Diffusion that brings familiar creative tools — layering, non-destructive editing and extensions — to diffusion workflows. With about 1,700 stars on GitHub, it aims to make AI image generation feel like working in a proper editor rather than typing into a bare prompt box.
Built as a portable front end that works with Stable Diffusion backends such as AUTOMATIC1111, OpenDream lets you stack generated and edited images in layers and adjust them non-destructively, preserving originals while you experiment — much like working in Photoshop. Its extension system is designed to be easy to write, so new operations can be added without forking the app, and sessions are portable so work can move between machines. The project grew out of frustration with prompt-box interfaces that discard your process between generations. Note that this OpenDream is varunshenoy’s diffusion editor — a distinct project from the other product of the same name listed on Tooloso.
Getting started is deliberately light: a single setup script brings up both the frontend and the backend, and the project documentation walks through connecting it to an existing Stable Diffusion installation.
OpenDream is free and open source under the MIT license. You run it on your own machine and connect it to your own Stable Diffusion setup, so there are no subscription fees from the project. The repository was last updated in 2023 and remains available as-is.
Stable Diffusion users who want an editor-like workflow with layers instead of one-shot generations, and creators who like extending their tools with small custom operations. It assumes you already run, or can run, a local diffusion backend such as AUTOMATIC1111.