---V1-Pony---
This is a model trained on the world of warcraft in-game images. It tries to recreate the same low poly character models.
I'm not gonna lie, this model takes a little tinkering to get working correctly and might be a bit over-cooked. I tend to run it on 0.5, but adjust as needed between 0.4~0.7
Necessary Activation Token:
WarcraftStyle
Additional tags that can help:
world of warcraft, low poly, 3d, blood elf, night elf, human, draenei,
Possible Negative tags:
blender \(medium\), blender, high poly
---V2-Pony---
Made changes to how 'over-cooked' the model was, and should now be able to be used between 0.8~1.0 rather comfortably without interfering with other lora as harshly.
Necessary Activation Token:
WarcraftStyle
Additional tags that can help:
world of warcraft, low poly, 3d, blood elf, night elf, human, draenei,
In testing, I think I've mostly reduced the necessity of Negatives for a decent final product, but try using the same negatives from V1 if you are still having issues.
---V1-Illustrious---
Model trained on same training data as V2-Pony, but for Illustrious.
Necessary Activation Token:
WarcraftStyle
Additional tags that can help:
world of warcraft, low poly, 3d, blood elf, night elf, human, draenei,
I've found that this model really likes the "low poly" tag, specifically to help with the unique armor style that wow tends to have. Adding this tag will generally help you get a more "wow-like" style.
If you like my work and want to give me some extra support, you can always tip⚡Buzz⚡to help me make more lora.
Made improvements to tag understanding.
Made improvements to it's 'over-cooked' nature, and be used much more liberally
Overall increase in quality and consistency
removed reliance on certain negatives to achieve desired look
1. Os direitos dos modelos repostados pertencem aos criadores originais.
2. Criadores originais que desejam reivindicar seus modelos devem contatar a equipe do SeaArt AI pelos canais oficiais. Clique para reivindicar
