Style Lora of 这也是QAQ - pixiv
Note: Every major version has a different style (datasets also do not overlap), so choose the one that best fits your needs.
Description
trained on Illustrious v0.1.
Interestingly, the overall color and details pop much more compared to the vpred version, even on vpred models.
It's possible that I've been doing the training wrong all this time.
Regardless, I feel like this is the peak for a while