This LoRA for Anima Preview 3 is trained using this Google Collab by @CitronLegacy. I used the default settings, chaning the number of Epochs to 8 and the number of repeats to 4. The whole run took about 4 hours.

The dataset 27 images focusing on Natsuki Subaru taken from anime screenshots and official art I took from ReZero's Wiki.

The result is better than I expected but worse than I hoped. This LoRA makes generating Subaru's eyes closer to the anime and way more consistent. Even with this LoRA, the model has tendency to give him feminine features, you must specify the 1boy tag.

THIS IS A PROOF OF CONCEPT VERSION and my First LoRA. Feedback Appreciated!