Blablabla SDXL | uin

Training size 1024*1024, there's not much explanation, just a style filter, the effect is referenced to the example image, the underlying model is recommended to be of the anime type,

Recommended starting resolution 800X1280.

Or start with a high-definition fix at 512*768,

Sampling Steps recommended 22-30, there's no need for too many step values that waste time processing the image;

CFG Scale recommended 6 to 8;

Weight: 0.7-1''

Welcome everyone to experience and use it, a newcomer who is always learning, many shortcomings please forgive!~~~~~~