About the hikigaya komachi tag
hikigaya komachi is a character tag that NovelAI Diffusion can use. On Danbooru it has 301 posts, and its best-rated still image scores 343. 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 hikigaya komachi
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 hikigaya komachi in a prompt
Put the subject first, then this tag, then quality tags.
hikigaya komachi, solo, best quality, masterpiecewlop, hikigaya komachi, 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 ▴
Character in Yahari Ore no Seishun LoveCome wa Machigatteiru. She is the younger sister of Hikigaya Hachiman.
She has black eyes, a fang/skin fang, small breasts, and short black hair with parted bangs and an ahoge. She usually wears a x hair ornament and a serafuku school uniform consisting of a green sailor collar, green neckerchief, white shirt, and a green pleated skirt. Later, the Sobu High School Uniform.
Voiced by: Yuuki Aoi (JPN), Hilary Haag (ENG)
FAQ
Does NovelAI recognize hikigaya komachi?
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 hikigaya komachi.
How do I write hikigaya komachi in a prompt?
Write "hikigaya komachi" 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.