Finetune of NoobAI, trained on a RTX 3060 for one full epoch and one partial, using this script. Dataset is about 4000 images of varied, anime adjacent art I manually curated and automatically captioned with WD EVA02-Large Tagger v3.

Text encoders were frozen and only the UNet was trained with conv1 and conv2 layers skipped. This helps enhance existing features without overfitting. Improvements are mostly backgrounds and artstyles.

As a base to train on, a merge of two existing finetunes was used; LS Tiro 3.0, Seele 2.1 and an unreleased version of NoobAI. I did a geometric sum on both of them against the base model with comfy-mecha. Then I added the Cosine A of both results with a ratio of 0.69. The VAE is finetuned; Luna. Thanks to Hysocs for the memory efficient RavenAdamW.