Style Collection

This is my personal style collection. Anything could be here, except artist style.

Cover images are directly from the vanilla (the original) base model in a1111, at default 1MP resolution. No upscale, no imprinting fixes.

All LoRAs in this collection, by default:

• Trained on illustrious v0.1.

• Prefer vanilla base model.

• Prefer vanilla base model.

• Prefer vanilla base model.

• Full natural language captions. No trigger words, just follow your heart cover images.

• Version means nothing, just my training job ID.

elpe v0.22:

• Dataset: Ultra-Ultra-Ultra-high-definition (~12K, no joke) renderings from the game engine. Mainly portraits. 100% homemade. Full of crazy lighting effects.

• Style: Ultra realistic textures and details, light partials, cinematic lighting effects, fire effects, etc...

• TBH, I did not believe SDXL can handle such complex dataset. But turns out it can, although not 100%, but looks cool enough.

• Support both 2D anime and 3D game characters. Prefer 3D characters for best effect.

dcr v0.7:

• Dataset: Mixed dataset mainly from the game Azur Lane.

• Style: Clean 2D anime illustrious, wallpaper.

• Purpose: a style regulator for vanilla base model. Giving you the feeling of the game style (vibe, colors, clothing, etc.).