LoRA is trained on images by artist Johanna Basford. LoRA was trained on a resolution of 768x768.

LoRA weight: I recommend using the LoRA weight between 0.7 and 0.85. I personally use 0.7.

Clip skip: The clip skip of the LoRA is 1.

For some reason sometimes LoRA doesn't activate, in which case I recommend adding "black and white" to the prompt.

I also recommend using "HiresFix".

Example prompt: A portrait of a anime girl with long hair and ponytail. All lines are perfect and smooth.