Stable Diffusion web UI Labeling extension – Tagger

What is Waifu Diffison 1.4 Tagger?

Image to text model created and maintained by MrSmilingWolf, which was used to train Waifu Diffusion.

Please ask the original author MrSmilingWolf#5991 for questions related to model or additional training.

Tagger for Automatic1111’s WebUI

Interrogate booru style tags for single or multiple image files using various models, such as DeepDanbooru.

한국어를 사용하시나요? 여기에 한국어 설명서가 있습니다!

Disclaimer

I didn’t make any models, and most of the code was heavily borrowed from the DeepDanbooru and MrSmillingWolf’s tagger.

Installation

1. Extensions -> Install from URL -> Enter URL of this repository -> Press Install button
   • or clone this repository under extensions/

$ git clone https://github.com/picobyte/stable-diffusion-webui-wd14-tagger.git extensions/tagger

2. (optional) Add interrogate model

• Waifu Diffusion 1.4 Tagger by MrSmilingWolfDownloads automatically from the HuggingFace repository the first time you run it.
• DeepDanbooru
  1. Various model files can be found below.
     • DeepDanbooru models
     • e621 model by 🐾Zack🐾#1984 (link contains NSFW contents!)
  2. Move the project folder containing the model and config to models/deepdanbooru
  3. The file structure should look like:

models/
└╴deepdanbooru/
  ├╴deepdanbooru-v3-20211112-sgd-e28/
  │ ├╴project.json
  │ └╴...
  │
  ├╴deepdanbooru-v4-20200814-sgd-e30/
  │ ├╴project.json
  │ └╴...
  │
  ├╴e621-v3-20221117-sgd-e32/
  │ ├╴project.json
  │ └╴...
  │
  ...

3. Start or restart the WebUI.

• or you can press refresh button after Interrogator dropdown box.
• “You must close stable diffusion completely after installation and re-run it!”

Model comparison

Model comparison

Screenshot

Artwork made by hecattaart

Copyright

Public domain, except borrowed parts (e.g. dbimutils.py)

Project Address