Danbooru Tags Transformer V2 with WD Tagger

Danbooru Tags Transformer V2 with WD Tagger

Labels an image with Danbooru-style tags and writes captions with Florence-2 models. Runs in the browser as a Gradio app, for people who label or caption anime-style pictures.

Made by John6666 on Hugging Face. These notes are Mindket's, written from its Space page and README; the app's own files and license are on Hugging Face.

About

What it is: This Gradio app works with Danbooru-style tags for images. It names tagger models that label a picture with tags, and Florence-2 models that write a caption for one. The Space runs on Hugging Face, so you can try it in a browser with nothing to install.

Who it suits: People who label, caption or prepare tags for anime-style images, and developers who want to see several tagger and caption models wired into one Gradio page.

Industry | Who uses it | How they use it | Example
Image generation | Prompt writers | Turn a picture into tags they can reuse as a prompt | Tag a reference image and read the tag list
Machine learning | Dataset builders | Label images with tags or captions | Run one image through a tagger and a captioner
Software | Developers | Study how app.py loads several models behind one page | Download the Space and read its files
Research | Students | Compare tagger models and caption models on the same image | Try one image with each tagger the Space names

Anyone who does not work with images or tags can skip it.

Where to use it: It runs in these places.

• A web browser, on the Space's page at Hugging Face
• Windows, with Python 3.12
• macOS, with Python 3.12
• Linux, with Python 3.12

Use case | Who | What they get back
Tag an image | Prompt writers and dataset builders | A list of Danbooru-style tags
Caption an image | Anyone with a picture to describe | A written caption from a Florence-2 model
Reuse the app code | Developers | A local copy they can read and edit

How to set it up: Using the Space online needs only a browser. A copy on your own computer needs the following.

• Python 3.12 and pip
• The huggingface_hub package, which gives you the hf command
• Disk space for the repository, which holds 25 files

1. Install the huggingface_hub package.
2. Download the Space's files into a local folder.
3. Move into that folder.
4. Install Gradio at the version the Space uses, then install the packages in requirements.txt.
5. Start the app.

pip install -U huggingface_hub
hf download John6666/danbooru-tags-transformer-v2-with-wd-tagger --repo-type space --local-dir danbooru-tags-transformer-v2-with-wd-tagger
cd danbooru-tags-transformer-v2-with-wd-tagger
pip install gradio==6.10.0
pip install -r requirements.txt
python app.py

How to use it: Open the Space's page on Hugging Face, or run the app on your computer and open the page Gradio shows. Then give it an image and read the tags and captions it returns.

python app.py

Tool | What it does
app.py | The main file, it builds the Gradio page
hf download | Copies the Space's files to your computer
requirements.txt | Lists the Python packages the app needs
pre-requirements.txt | A second package list in the repository, left out of the steps above

Example: Open the Space in your browser and give it one anime-style picture. You get back Danbooru-style tags for what the picture shows, and you can ask a Florence-2 model for a written caption of the same picture.

For newcomers and experts: A tag is a short label for something in a picture, such as an object or a style. The first small thing to try is the online Space with a single image.

The Space names six models. Three are taggers, one WD EVA02 tagger and two from the idolsankaku family in EVA02 and SwinV2 builds. Three are Florence-2 captioners, Flux, Flux-Large and SD3-Captioner. The hosted copy is set to run on Hugging Face hardware called zero-a10g.

You give the page an image
A tagger model labels it with tags
A Florence-2 model writes a caption
The page shows the tags and caption

Good to know: The Space states no usage limits, so treat it as a tool you can read and adapt.

• License: apache-2
• Cost to run: none named
• Data it reaches: the image you give it, plus the Space files and model files that Hugging Face supplies
• Hardware: the Space is set to zero-a10g at Hugging Face, and what your own computer needs is not stated

On Hugging Face, 13 November 2026

• Likes: 325
• Built with: gradio
• License: apache-2.0

Links

• Space: https://huggingface.co/spaces/John6666/danbooru-tags-transformer-v2-with-wd-tagger
• Files: https://huggingface.co/spaces/John6666/danbooru-tags-transformer-v2-with-wd-tagger/tree/main

Try it

Open the Space in your browser, or copy it to your own Hugging Face account from its page. Nothing in this kit runs by itself.