V1.0 uses a certain private model as the pre-training model for training, with a resolution of 1024*1024.
V2.0 uses nai as the pre-training model for training, with a resolution of 768*768
Trigger:qianzhi, hair ornament, red eyes, japanese clothes, brown hair, kimono, gloves, bangs, black gloves, multicolored hair, hair flower, side ponytail, elbow gloves, pantyhose