Diffusion Bee is a macOS GUI application for running Stable Diffusion locally, providing one-click installers and offline image generation. It targets users who require privacy and local execution, and includes features like inpainting, ControlNet support, upscaling, and model management.
Key Features
- Local generation: runs inference on the user’s machine, with no cloud upload by default.
- One-click installer and user-friendly UI for non-technical users.
- Support for multiple models and features: SD 1.x, SD 2.x, SD XL, Inpainting, ControlNet, LoRA.
- Performance optimizations for Apple Silicon (M1/M2).
Use Cases
- Local creative image generation with data privacy.
- Rapid experimentation with prompts and models for art and product images.
- Offline or edge deployments where cloud access is limited or undesired.
Technical Highlights
- Electron-based desktop UI with integrated model management and backend tooling.
- Model downloading and user-managed model/plugin support.
- Distributed under AGPL-3.0 license; note that model licenses (e.g., Stable Diffusion weights) may impose separate terms.