Lora of Noelle Silva from Black Clover. Ive been experimenting with lora training on anima and so i decided to try something new on noelle. Ive scraped around 12k screenshots from fancaps, filtered it down to 313 images, upscaled them and used wd14 tagger. After that ive changed captions for different outfits to outfit 1, outfit 2 etc. so it binds to a short prompt. Its not perfect but it works i guess ¯\_(ツ)_/¯. I always wondered how @MetalchromeX does it. 1280 x 2048 works greatBasic prompts :<ul><li>noelle silva</li><li>outfit 1 (classic outfit with capelet on)</li><li>outfit 2 (no capelet)</li><li>valkyrie, valkyrie outfit (you can add polearm, holding polearm for weapon etc.)</li><li>maid</li><li>bikini</li><li>hair bun (for the onsen scenes)</li></ul>

Files with SHA256: 833b2684b03ecf9b10eb8f2fd3872dbefeb1fda0f9f26306e65522d012c8c3e8