
This is my first SDXL LoRA, so please be nice.
This LoRA makes images with a gothic chibi doll look and should work with most SDXL models.
Usage Guide
-
While this doesn't require a trigger word, adding 'gothic doll' somewhere in the prompt can help.
-
A LoRA strength between 0.5 to 1.5 is recommended.
-
Higher strengths than 1 will increase the style strength but can introduce visual artifacts.
-
When up-scaling it can help to reduce the LoRA strength to allow the base model to fix some of the artifacts introduced.
This was trained off a dataset of my own images, so if people like it I'll make a flux version.
描述:
训练词语:
名称: 952832_training_data.zip
大小 (KB): 4940
类型: Training Data
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: ZD-Gothic-Chibi-Dolls.safetensors
大小 (KB): 223102
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success