DANBOORU TAGGING SYSTEM
=======================

What's Danbooru?
----------------

Big anime image board that's been cataloging and tagging artwork since
2005. Every image gets standardized tags - character details, art style,
composition, everything.

When people trained AI models on this data, the models learned to connect
specific tags with visual stuff. Those tags basically became a programming
language for making images.


Why This Matters
----------------

Danbooru tags are precise. Instead of writing:

  "a young woman with long flowing blonde hair wearing a red dress"

You break it into chunks the model actually recognizes:

  1girl, long hair, blonde hair, red dress

Each tag maps to something the model learned. It combines them. Way more
control than natural language.


Tag Types
---------

Character count:    1girl, 1boy, 2girls, solo, multiple girls
Physical traits:    blonde hair, blue eyes, tall, muscular
Clothing:           dress, suit, armor, school uniform
Poses:              standing, sitting, arms crossed, looking at viewer
Expressions:        smile, serious, crying, surprised
Backgrounds:        outdoors, classroom, forest, simple background
Meta/quality:       masterpiece, best quality, highres


Tags vs Plain English
---------------------

Some models (especially photorealistic ones) like natural language better.
But anime models and most SDXL checkpoints expect tags.

Mix and match works too:

  1girl, detective, woman in her 30s, noir style, standing in rain

Model grabs what it knows and figures out the rest from context.


Order Matters
-------------

Tags up front carry more weight.

  1. Quality tags first - sets the baseline
  2. Character count early - establishes who's in the shot
  3. Physical stuff - general (age, skin) to specific (hair, eyes)
  4. Clothing, expression, pose come after

Tags right after a character description bind to that character.