divamgupta/diffusionbee-stable-diffusion-ui

Diffusion Bee is a desktop application for macOS that lets you generate AI images on your own computer without sending data to an external service. It provides a graphical interface for Stable Diffusion, an open-source AI image generation system. The application is designed for non-technical users: installation is a single download with no dependencies, no command-line tools, and no configuration required. Once installed, you type a text description of the image you want to generate, click a button, and the application produces an image. All processing happens locally on your Mac using the machine's own hardware. No account is required, and no image data is sent to the cloud unless you explicitly choose to upload a reference image. The application supports several generation modes. Text-to-image creates a new image from a text description. Image-to-image takes an existing image as a starting point and modifies it based on a text prompt. Inpainting lets you paint over a region of an existing image and replace just that area with AI-generated content. Outpainting extends an image beyond its original borders. ControlNet, a technique for constraining the generated image to follow the structure of a reference image such as a pose or edge map, is also supported. LoRA files, which adjust the visual style of the generated output, can be used as well. Supported model families include Stable Diffusion 1.x, 2.x, and XL. Models can be downloaded directly from within the application. There is a generation history so you can review past outputs, plus an upscaling tool to increase the resolution of results. The app is optimized for Apple M1 and M2 chips but also runs on Intel Macs. The operating system requirement is macOS 12.3.1 or later for Intel, and macOS 11.0 or later for M1 and M2 machines.