AIYA

A Discord bot interface for Stable Diffusion

Setup requirements

• Set up AUTOMATIC1111's Stable Diffusion AI Web UI.
  • AIYA is currently tested on commit bb2e2c82ce886843f11339571f9a70d4c5f2a09d of the Web UI.
• Run the Web UI as local host with api (COMMANDLINE_ARGS= --listen --api).
• Clone this repo.
• Create a text file in your cloned repo called ".env", formatted like so:

# .env
TOKEN = put your bot token here

• Run the AIYA by running launch.bat (or launch.sh for Linux)

Usage

To generate an image from text, use the /draw command and include your prompt as the query.

Currently supported options

• negative prompts
• swap model/checkpoint (see Notes or wiki)
• sampling steps
• height/width (up to 768)
• CFG scale
• sampling method
• seed
• img2img
• denoising strength
• batch count
• Web UI styles
• face restoration

Bonus features

• /settings command - set per-server defaults for the following (also see Notes!):
  • negative prompts
  • model/checkpoint
  • sampling steps / max steps
  • sampling method
  • batch count / max batch count
• /identify command - create a caption for your image.
• /stats command - shows how many /draw commands have been used.
• /tips command - basic tips for writing prompts.
• /upscale command - resize your image.

Notes

• Ensure AIYA has bot and application.commands scopes when inviting to your Discord server, and intents are enabled.
• As /settings can be abused, consider reviewing who can access the command. This can be done through Apps -> Integrations in your Server Settings.
• React to generated images with ❌ to delete them.
• Optional .env variables you can set:

URL = set URL if yours is not default (http://127.0.0.1:7860)
DIR = set folder to save images, otherwise the default is \outputs

# only set USER and PASS if you use --share and --gradio-auth

USER = your username
PASS = your password
COPY = set to anything to have the output show the command used to produce the image instead of the prompt

• On first launch, AIYA will generate a models.csv with a default dummy value. If you'd like to add more models/checkpoints, replace the default value and add lines following the header format.
  • Display name is anything you want.
  • Model full name is how it would appear in the Web UI.
  • Activator token is any words a model may need to take effect. Can be blank (remember to add | separator at the end).
  • An example may look like:

display_name|model_full_name|activator_token
SD 1.5|v1-5-pruned-emaonly.ckpt [81761151]|
WD 1.3|wd-v1-3-float32.ckpt [4470c325]|
MoDi 1|moDi-v1-pruned.ckpt [ccf3615f]|Modern Disney style

• In the Web UI, there is a setting named "Checkpoints to cache in RAM". If you have enough RAM, this value can be increased to speed up swapping.

Credits

AIYA only exists thanks to these awesome people:

• AUTOMATIC1111, and all the contributors to the Web UI repo.
  • https://github.com/AUTOMATIC1111/stable-diffusion-webui
• harubaru, my entryway into Stable Diffusion (with Waifu Diffusion) and foundation for the AIYA Discord bot.
  • https://github.com/harubaru/waifu-diffusion
  • https://github.com/harubaru/discord-stable-diffusion
• gingivere0, for PayloadFormatter class for the original API. Without that, I'd have given up from the start. Also has a great Discord bot as a no-slash-command alternative.
  • https://github.com/gingivere0/dalebot
• You, for using AIYA and contributing with PRs, bug reports, feedback, and more!