Well the time I upload isn't that good huh...This is my first ever lora model, you can check out my Pixiv for more previews.<a target="_blank" rel="ugc" href="https://www.pixiv.net/users/18763851">4BEraser - pixiv</a>My version is focus on the official outfit of C.B, with たろうまる's facial expression &amp; art style. The result is better than other one's version but sometimes it can't recognize the single sleeve of chara.I don't know what is best, but I do usually use a mix model of Counterfeit &amp; PastelMix, other anime model (Anything, Orange) are compatible.How to use:You can input nothing to generate the character any way, but I do recommend you to use following prompts to generate the outfit:<ul><li>Base: umamusume, green bow, green eyes, horse ears, horse girl, ahoge, bow, hat bow, hairclip, long hair, horse tail, hair ornament,</li><li>Outfit: (white pants:1.2), midriff, navel, bangs, jacket, belt, flare pants, strapless shirt, mini top hat, green shirt, (single sleeve1.4), crop top, white jacket, arm strap, open clothes, open jacket,<ul><li>It isn't designed for alternative outfits, but you can put the Outfit tag in negative prompts for a try, I have put some examples in previews.</li><li>If you use base prompt only it will be tempting to generate portrait, if so, please add upper body or full body tag.</li><li>If you find out that it's focusing on chara's feet, you can try to use Controlnet to limit it.</li></ul></li></ul>Settings are what I am using, you can change by your own.<ul><li>DPM++ Series(SDE Karras, 2M Karras, etc.)</li><li>about 20 steps, CFG scale 5~7</li><li>LoRA Weight: 0.6-0.8</li><li>If using highres<ul><li>ESRGAN_4X (Not good at portrait)</li><li>Denoising 0.5-0.6.</li><li>P.S. LoRA training resolution is 768*512.</li></ul></li></ul>

Files with SHA256: c1b8d37013a433f7e9276d7d02a15f5742c840e7baf979cb1e202b558db1369b