Trained on Tsubaki, 13 images.
DiT V3: https://pixai.art/en/model/2057708310480581621
Hands are a bit worse compared to Rias LoRA, but it's not bad. I didn't have to burn credits to get a good result, like with previous model types like SD and SDXL. As you can read, this LoRA is specifically for Xenovia in her battle suit. If you can pull of her school uniform, which I have not used at all, then I'll be really impressed. Also, her anime sword won't generate, which I expected. If you can make LoRA for her sword, I think it would work. Her outfit isn't 100% accurate, which is fine, I guess.
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. It's a change from the previous SDXL and SD models.
LoRAs I used in generation: https://pixai.art/en/model/1839426804048327588
Model used in thumbnail: Tsubaki
