About the clift tag
clift is a character tag that NovelAI Diffusion can use. On Danbooru it has 256 posts, and its best-rated still image scores 183. 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 clift
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 clift in a prompt
Put the subject first, then this tag, then quality tags.
clift, solo, best quality, masterpiecewlop, clift, 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
A character in Dragon Quest IV. Along with Brey, he is one of the companions of Alena and usually drawn in connection with her.
His hat is a mitre.
In the original NES localization, he was named Cristo while in modern localizations (from the DS remake onward), he is named Kiryl and he speaks with a thick Russian accent just like his companions whenever he has dialogue.
He is infamous in the original versions of the game for his AI, where he will always try to cast his instant kill spells first on bosses (who are likely to be immune in the first place) instead of doing his job of healing the party.
FAQ
Does NovelAI recognize clift?
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 clift.
How do I write clift in a prompt?
Write "clift" 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.