Trigger Word: @QAQv1 ~ @QAQv10

This is an all-in-one LoRA featuring 10 different styles from pixiv QAQ. (Experimental)

The styles correspond to v1 through v10 in the order of the sample images.

Dataset: 1,000 images in total (100 images per style).

Future Improvements:

• Confirmed that some styles fit better than others. Accordingly, the dataset size needs to be adjusted and balanced for each style.

• A higher learning rate is required to achieve better reproducibility and fidelity.