Trained on Tsubaki, 13 images.
The "Bomb Queen". Actually surprised to see this LoRA work so well. There is a lot of detail on Yubelluna clothing, and because there is a relatively strict letter limit on LoRAs here on PixAI, I wasn't able to include all the details, like her purple lipstick, which you can manually add during the generation or her " forehead thingy". I cannot be lazy and simply slap "school uniform" on Yubelluna. Generated images are better than the "regular version" of Xenovia, even though the image quality was worse in Yubelluna's case, as she only appears in season 1, in my opinion.
I only used Tsubaki to generate the images, but I think you can use other DiT models.
I use Euler A. 28 steps. Because the results are pretty good, I think you can generate 3 images in parallel, and you will still be fine. Unlike SDXL, DiT is pretty consistent from my initial experience. Of course, if you demand a lot of detail, then maybe go cheap first.
I use 3, 4.5 and 6CFG. 0.7 Rescale CFG, same as every DiT V1 LoRAs I use.
LoRAs I used in generation: https://pixai.art/en/model/1839426804048327588
Thumbnail model: Tsubaki
