Natural Text to Danbooru Tags with Transformer V2

Natural Text to Danbooru Tags with Transformer V2

Turns a plain-language description into Danbooru-style tags using language and tagger models. Runs in the browser as a Gradio app, for people who write tag-style prompts for images.

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 converts a description in ordinary words into Danbooru-style tags. It names several language models and tagger models for the job, and its repository ships tag lists as CSV and text files. The Space runs on Hugging Face, so you can try it in a browser with nothing to install.

Who it suits: People who write tag-style prompts for anime-style images, and developers who want to see how tag lists and language models fit into one Gradio page.

Industry | Who uses it | How they use it | Example
Image generation | Prompt writers | Turn a sentence into tags in the Danbooru style | Describe a scene in plain words and read the tags
Machine learning | Dataset builders | Prepare tag text to go with images | Convert a batch of short descriptions into tag lines
Software | Developers | Study how the Python files and tag lists work together | Download the Space and read tagger.py and tags.txt
Research | Students | Compare the language models the Space names | Try the same description with each model

Anyone who does not use Danbooru-style 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
• macOS, with Python
• Linux, with Python

Use case | Who | What they get back
Convert text to tags | Prompt writers | A set of Danbooru-style tags for the description
Tag from an image | Dataset builders | Tags from the tagger model the Space names
Reuse the code and tag lists | 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 and pip
• The huggingface_hub package, which gives you the hf command
• Disk space for the repository, which holds 16 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/text2tags-transformer --repo-type space --local-dir text2tags-transformer
cd text2tags-transformer
pip install gradio==4.37.2
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 type a description in plain words and read the tags 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
tags.txt | A tag list stored in the repository

Example: Open the Space in your browser and type a short description of a character in a garden. You get back a line of Danbooru-style tags that match the description, ready to reuse as a prompt.

For newcomers and experts: A tag is a short keyword for something in a picture. This app writes those keywords for you from a normal sentence. The first small thing to try is the online Space with one short description.

The Space names five models, among them two dart-v2 builds, a SwinV2 tagger and two llama-tagger builds, one of them a 4-bit GPTQ version. The repository also holds CSV tag groups and character lists. The Space is built on Gradio 4.37.2, an older version than most apps use today.

You type a description
A language model proposes tags
Tag lists and tagger code refine them
The page shows the tag line

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

• License: openrail
• Cost to run: none named
• Data it reaches: the text you type, plus the Space files and model files that Hugging Face supplies
• Hardware: no hardware is named, and what your own computer needs is not stated

On Hugging Face, 13 November 2026

• Likes: 16
• Built with: gradio
• License: openrail

Links

• Space: https://huggingface.co/spaces/John6666/text2tags-transformer
• Files: https://huggingface.co/spaces/John6666/text2tags-transformer/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.