All text translated into English using DeepL :-(

Discovered the surprising thing that PonyXL is missing a LORa on Alexstrasza from World of Warcraft. I decided to correct this injustice. For the dataset, 142 images were taken from rule34.

However, I'm still new to training and don't understand how it works at all. So I just follow other people's guides.

This LORa is trained on Autismmix_confetti.

Words Trigger: AlexstraszaCute

Support words: red hair, long hair, orange eyes, horns

Armor tag: bikini armor, shoulder armor

But you are free to change hair length, chest size, etc.

All examples have strength = 0.9

I noticed the following problems:

1. sometimes gives out VERY bad hands.

2. sometimes duplicates the pupil.

p.s: Please note that lora's name in my examples is different from the posted name. I renamed the file to something "logical" for some convenience.
p.p.s: The examples show 4 pictures that are created without unnecessary corrections. Only hires fix. This is done so that you can see the REAL result that LORa produces. The fifth picture is already using third-party LORa, Hires Fix, Adetailer, Photoshop, etc.