To tell you the truth, I don't really like this character or the manhua; I simply happened to have collected these training images before. I tried using Gemini 3.0 Pro to help me tag all the images, thinking it would save time, so I didn't double-check them. After training, I realized some of the tags were completely wrong.

Personally, I'm not entirely satisfied with this LoRA. However, since I spent Buzz to train it, I decided to publish it anyway. It's still a piece of work, right? Hopefully, some of you will still find it useful!

ATTENTION: DIGITAL WATERMARK ACTIVE

This LoRA has been injected with unique weight-level identifiers. While it does not affect image generation, it allows for permanent traceability of the model's lineage.

Any unauthorized re-hosting or commercial usage will be identified via automated weight-matching scripts. Respect the creator; do not re-upload or resell.