Gauss is an open-source native macOS app for Stable Diffusion image generation, built with SwiftUI on top of Apple’s ml-stable-diffusion CoreML models. Hosted on GitHub with about 1,300 stars under the GPL-3.0 license, it runs diffusion models entirely on your Mac — Apple Silicon is recommended — with the last code updates in 2023.
Unlike web interfaces or Python-based tools, Gauss is a document-based Mac application: you create a document, write a prompt in the composer at the bottom of the window, adjust parameters, and generate. Every image you create is stored locally inside the document’s .gaussnb notebook file, so your work stays organized and private on your own machine. On first use you install Stable Diffusion models (each around 2.5 GB) that run through CoreML, which lets the app take advantage of Apple’s Neural Engine for fast on-device generation. Finished images can be dragged straight into Finder or any image editor. The author describes the software as alpha quality and under development, and it requires macOS 13.1 or later.
Gauss is free and open source under the GPL-3.0 license. Builds are available from the GitHub releases page at no cost, and because generation happens on your own Mac, there are no credits, subscriptions or per-image fees.
Mac users who want a simple, private, native way to generate Stable Diffusion images on their own hardware, especially Apple Silicon owners who prefer a document-based app over Python web UIs.