This is an attempt to port the "folded ponytail" tag from Illustrious (which is rock solid at it*) to Z-Image (which has no idea**). Since calling it by its Danbooru name would lead to collisions with actual ponytails, I've called it "folded updo" here.
Note: Since this has a considerable amount of anime-ish images in the dataset, you may have to prompt for realism/photography more explicitly than you're used to. Conversely, if you prompt for anime/illustration and not for the hairstyle, this actually works as a decent "generic anime" style LoRA.
* all Danbooru-based models are, likely thanks to Kaguya Shinomiya
** Same for Flux and Qwen, which I might make later. Chroma can do it already.
Initial version created from 36 images made by various Illustrious models with different styles (realistic ones with Hyphoria Real, others I've tried have lost much of the concept)
1. Права на репостнутые модели принадлежат оригинальным создателям.
2. Для подтверждения авторства модели создателям следует обратиться к сотрудникам SeaArt AI через официальные каналы. Нажмите, чтобы подтвердить
