Skip to content
forked from Efreak/stable-ui

🔥 A frontend for generating images with Stable Diffusion through Stable Horde

License

Notifications You must be signed in to change notification settings

rbrtcs1/stable-ui

 
 

Repository files navigation

fork notes

THIS FORK WILL NOT BE MAINTAINED OR UPDATED.

I suggest using ArtBot instead 😏

This is a fork of aqualxx/stable-ui, an unmaintained frontend for AI Horde crowd-sourced stable diffusion. It contains all the same bugs and limitations the original stable-ui contains, but has been updated to support SDXL, which currently requires a batch size of 2, and recommends a minimum of 1024x1024. Instead of updating this properly, I've simply added a check to see if the model includes SDXL, and changed the requested number of images to 2; that means for every 1 SDXL image you request here, you'll get 2 images back instead of 1. The only niceties I've added beyond this is to set the default model to SDXL and the defaults size to 1024x1024. Everything else is the same.

Update: I applied a fix to the leaderboard, as found by @super.skirv on Discord.

Update: I applied a fix to space out requests because horde now has a rate limit on async endpoints. Since Horde now supports SDXL officially, I also swapped out sdxl_beta model name for SDXL 1.0 and removed the inpainting and image amount overrides from it.

Jan. 8, 2024: Added LoRa/TI support. In your prompt, use the syntax <lora:id:strength> or <ti:id> to add them to your request. For LoRas, put a 'v' before the id to use the version id instead. Not that for both loras and tis, only integer ids are supported. (Added by rbrtcs1)

Update: I added LCM sampler. TODO: if upcoming worker changes allow it, add unipc sampler and allow use of dpm samplers in img2img. -Efreak

stable-ui 🔥

Stable UI is a web user interface designed to generate, save, and view images using Stable Diffusion, with the goal being able to provide Stable Diffusion to anyone for 100% free.

This is achieved using Stable Horde, a crowdsourced distributed cluster of Stable Diffusion workers, which makes this tool available for anyone to use regardless if they don't have the processing power to run Stable Diffusion themselves.

Features

  • 🎨 Image generation utilizing Stable Horde
    • No installation required!
  • ⚙️ Tons of customizable settings
    • Support for Text2Img, Img2Img, and Inpainting
    • Adjust settings for resolution, batch size, NSFW, etc
    • Employ over 50 different Stable Diffusion models!
  • 🖼️ A gallery to view all generated images
    • Favourite, download, delete, and share images
    • Reuse image settings for Text2Img or send back an image for Img2Img or inpainting
    • Use batch selection to perform actions on multiple images at a time
    • Store images locally through the browser
  • 📰 A dashboard for displaying Stable Horde info
    • View horde/user statistics
    • Edit your Stable Horde workers
    • See user rankings through a leaderboard
    • Read updates through a news feed
  • 📜 Statistics for each available Stable Horde worker, team, or model
    • Sort by parameter or filter by name
  • 💻 Optimized designs for both desktop and mobile devices

Contributing

Feel free to contribute! Pull requests are welcome.

Special Thanks To

Library Description
Vue 3 An amazing Javascript framework
Pinia For creating Vue stores
Element Plus Beautiful Vue 3 components
VueUse Tons of useful hooks
XIcons An extensive library of icons

License

MIT License

About

🔥 A frontend for generating images with Stable Diffusion through Stable Horde

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Vue 52.5%
  • TypeScript 45.7%
  • Other 1.8%