automatic1111/stable-diffusion-webui

Stable Diffusion web UI is a web interface for Stable Diffusion, the open-source AI model that turns text prompts into images. The README opens by saying the project is implemented using the Gradio library, which is a Python toolkit for putting a browser-based form in front of a machine-learning model. Instead of editing scripts and calling the model from a terminal, you fill in a prompt, pick options, and hit Generate. The features the README lists give a sense of what it covers. There are the original txt2img and img2img modes, plus outpainting, inpainting, color sketch, prompt matrix, an Extras tab for face restoration and upscaling using tools like GFPGAN, CodeFormer, RealESRGAN, ESRGAN, SwinIR and LDSR, a Checkpoint Merger that combines up to three model checkpoints, an X/Y/Z plot for comparing parameters, textual inversion training, hypernetworks, Loras, attention syntax that lets you weight parts of a prompt, negative prompts, styles, prompt editing mid-generation, batch processing, a built-in API, and many community extensions through custom scripts. It supports Stable Diffusion 2.0 and Alt-Diffusion, loads checkpoints in safetensors format, and works on as little as 4GB of video memory according to the README. You would actually use it if you want to generate or edit images locally on your own machine and you want a friendlier interface than the command line. The installation section lists guides for NVidia, AMD, Intel and Ascend NPU hardware, plus online services like Google Colab if you do not have a suitable GPU. The tech stack is Python with Gradio, and the project is centered on running Stable Diffusion locally.