Stable Diffusion web UI
A feature-packed browser interface for generating and editing images locally using Stable Diffusion models.
Generate, inpaint, upscale, and fine-tune AI imagery using a local browser dashboard and custom models.
Stable Diffusion WebUI gives you a web interface powered by the Gradio library to generate and edit images on your own computer. It is built for creative professionals and digital artists who want detailed control over image generation parameters without relying on paid cloud platforms.
What it is
The application provides tools for text-to-image, image-to-image, inpainting, outpainting, and face restoration using GFPGAN and CodeFormer. It includes visual workflow features like prompt attention weighting, checkpoint merging, embedded image metadata extraction, and custom extension support.
Who it's for
This tool suits graphic designers, illustrators, and digital artists who have a dedicated GPU and want precise control over AI generation workflows. It is not suitable for users without dedicated graphics hardware or those who prefer zero-setup managed web applications.
Key features
Inpainting and Outpainting
Modify designated regions of an image or extend existing canvas borders using text prompts.
Prompt Weighting and Matrix
Adjust keyword emphasis using simple syntax and test prompt combinations across matrix grids.
Model and LoRA Management
Merge up to three checkpoints and apply custom LoRAs, hypernetworks, or embeddings directly within prompts.
Integrated Face Restoration
Fix facial details and upscale low-resolution outputs using integrated models like GFPGAN, CodeFormer, and RealESRGAN.
Metadata Saving and Retrieval
Store prompt and sampler settings inside output PNG chunks, then drag images back into the UI to reload parameters.
Up and running in 3 steps
Our honest verdict
What's great
- Reads and automatically restores full generation parameters embedded in PNG and JPEG image files.
- Supports community extensions, custom scripts, hypernetworks, and LoRAs directly inside the interface.
- Includes specialized face restoration neural networks like GFPGAN and CodeFormer alongside multiple upscalers.
- Offers fine prompt control through attention weighting, mid-generation prompt editing, and composable prompts.
Where it falls short
- Requires local hardware with a dedicated GPU (at least 4GB VRAM recommended) to run effectively.
- Requires technical installation steps including setting up Python 3.10.6, Git, and running local execution scripts.
- Lacks built-in cloud hosting out of the box, requiring you to configure services like Google Colab if local VRAM is lacking.
Pricing & license
The software is free to use under the AGPL-3.0 licence. Your only expenses come from the hardware needed to host it locally or third-party cloud compute fees if you choose to run it on external servers.
Open remote AI-training roles
870 open remote AI-training roles, updated November 14, 2026; most pay $20–$170 an hour. Zekai lists them free; the application goes to the employer’s platform.
Alternatives
These reviewed AI tools do the same job instantly — from our scored directory.
Use it, or have us set it up
FAQ
What hardware do you need to run this locally?
You need an NVIDIA, AMD, or Intel GPU with at least 4GB of VRAM, though low-memory configurations can run with 2GB.
How do you recover generation settings from an image you made?
Drag and drop the generated PNG file into the PNG Info tab to view and copy the exact prompt, seed, and sampler details.
