Stable Diffusion Open Source Projects
Stable Diffusion is a family of generative image models that creates or transforms images through an iterative denoising process. It can turn text prompts into images, modify existing images, and support tasks such as inpainting and upscaling. These capabilities help artists, developers, and researchers explore visual ideas, automate image workflows, and adapt models to particular styles or use cases.
Open source tools in this area include image-generation interfaces, workflow editors, model training utilities, extensions, and APIs. When choosing a tool, check its license and model requirements, hardware and software compatibility, maintenance activity, documentation, and integration options. Consider how much control you need over generation and whether the tool supports your preferred models and workflows. These projects can be useful to creators, researchers, and developers building or running image-generation systems locally or as part of larger applications.
3 repositories · updated September 21, 2026

Guaardvark: Your Self-Hosted AI Studio for Agents, Media, and Code
Guaardvark is a comprehensive, self-hosted AI studio designed for local execution of advanced AI tasks. It integrates coding agents, media generation (video, image, music, voice), and robust RAG capabilities, all running on a single GPU. This platform prioritizes privacy and user control, enabling a full AI workstation experience on your own hardware.

AUTOMATIC1111/stable-diffusion-webui: Powerful AI Image Generation Web UI
The AUTOMATIC1111/stable-diffusion-webui project offers a comprehensive web interface for Stable Diffusion, simplifying AI art generation. It provides a robust set of features, including text-to-image, image-to-image, inpainting, and upscaling, all within a user-friendly environment. This Python-based UI is a popular choice for both beginners and advanced users exploring generative AI.

PPS-Ctrl: Controllable Sim-to-Real Translation for Colonoscopy Depth Estimation
PPS-Ctrl is an innovative image translation framework designed for controllable sim-to-real translation in colonoscopy depth estimation. It leverages Stable Diffusion and ControlNet, guided by a unique Per-Pixel Shading (PPS) map. This approach provides a physics-informed structural prior, enhancing texture realism and structure preservation in medical imaging applications.