Тип
LORA
Время публикации
2023-03-20
Базовая модель
SD 1.5
Версия введения.
More diverse training set; doubled network dimension; trained at 768 resolution. Better ??????? and composition than v2.
Trained on 2267 mostly ???? bara/yaoi images. Images are tagged with wd-v1-4-swinv2-tagger-v2.

