About the kirakishou tag
kirakishou is a character tag that NovelAI Diffusion can use. On Danbooru it has 603 posts, and its best-rated still image scores 99. The samples above put the tag at the front of a fixed base prompt with seed 1234, so the differences between tags come from the tag alone.
How each model draws kirakishou
Follows composition and expression instructions closely, so the effect shows in a tidy form.
Reacts to the tag more strongly, so its effect is easier to see.
WAI illustrious SDXL v17, a local model. Shows how the same tag reads outside NovelAI.
Writing kirakishou in a prompt
Put the subject first, then this tag, then quality tags.
kirakishou, solo, best quality, masterpiecewlop, kirakishou, soloThe negative prompt and the full settings are on the terms page.
How these samples were made
Every image uses 832×1216, seed 1234, 23 steps and the k_euler_ancestral sampler. The base prompt and negative prompt stay the same for every tag, and only the tag under test changes. Images flagged by the NSFW check are regenerated with a new seed or hidden.
From the Danbooru wiki
Show the wiki text ▾Hide ▴
Snow Crystal
The seventh and final Rozen Maiden doll. She is both emotionally and physically unstable, being the only doll not given a body. She fights using white rose vines. She does not have a master, instead choosing to kidnap the others' masters.
She has very long wavy blonde hair with two side up held with roses and sidelocks and yellow eyes with a white rose over one eye. Her outfit is similar to Barasuishou's lolita fashion, consisting of a frilled white shirt with wide juliet sleeves, a frilled white skirt, and white lace-up high-heel thigh boots.
Appearance
FAQ
Does NovelAI recognize kirakishou?
Compare the samples on this page. Every tag on this site shares the same seed and base prompt, so any change you see comes from kirakishou.
How do I write kirakishou in a prompt?
Write "kirakishou" near the front of the prompt and separate it from other tags with commas.
Should I use V5 or V4.5?
V4.5 shows the tag's effect more strongly. V5 gives steadier images with fewer artifacts.