Trained on Tsubaki, 13 images.
Applies for DiT V2: Use these prompts to get more accurate results for her: Yasaka, High School DxD HERO, long hair, adorned with black hair comb ornaments featuring gold markings, blonde hair, yellow eyes, ponytail, yellow kimono, a golden obi white robe belt, black Haori with gold skulls and lines, cleavage, huge breasts,
Whether you liked or disliked Season 4's different art style, there is one busty maiden everyone enjoyed, Yasaka. Hope you enjoy. I could have tried to get her hairpins done correctly, but that would have got me over the letter limit, and I know it would still most likely generate inaccurate results.
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.
Thumbnail model: Tsubaki
