As the title says, its the mom (Touka), from Suiren no Yado by Asakawa.I just wanted to see her colored instead of black and white, and I'm quite happy with how it turned out. xDThe character was trained on Hoseki - LustrousMix [IllustriousXL] (so use that to create)Tags to Use: sdnaoshi (for activation)Other Tags: 1girl, large breasts, very long hair, black hair, green eyes, earrings, jewelry, hair over one eye, eye visible through hairInstead of green eyes you can use aqua eyes too, and you can remove the hair over one eye if you don't want that. As for other tags, just change it based on what you want, like add comic and what not.My go-to negative tags: bad quality, worst quality, worst detail, lowres, bad anatomy, signature, watermark, greyscale, monochrome, censored, pixelated, mosaic, bar censor, text, speech bubble I often do Character LORAs from webtoon/manhwas. I prefer if the character looks the same or as close to the original as possible, so I may have overtrained my Loras. I hope you will all like it though. Please consider donating on if you like it, and give me tips if you want. <a target="_blank" rel="ugc" href="https://ko-fi.com/feridian">https://ko-fi.com/feridian</a> 白黒ではなくカラーで見てみたかったので、仕上がりにはとても満足していますキャラクターは Hoseki - LustrousMix [IllustriousXL] でトレーニングしました。（なので、それを使って作成してください）使用するタグ: sdnaoshiその他のタグ: 1girl, large breasts, very long hair, black hair, green eyes, earrings, jewelry, hair over one eye, eye visible through hair緑の目の代わりにアクア色の目を使ってもOKですし、片目を隠す髪も必要なければ外して大丈夫です。他のタグも、好みに合わせて変更してください。例えば、コミック風にしたいなら追加するといいと思います。よく使うネガティブタグ: bad quality, worst quality, worst detail, lowres, bad anatomy, signature, watermark, greyscale, monochrome, censored, pixelated, mosaic, bar censor, text, speech bubble普段、WebtoonやマンガのキャラクターのLoRAを作成しています。できるだけオリジナルのキャラクターに近づけたいので、LoRAを少し過学習させてしまっているかもしれません。それでも気に入ってもらえたら嬉しいです。もし気に入っていただけたら、寄付やアドバイスをいただけると嬉しいです。<a target="_blank" rel="ugc" href="https://ko-fi.com/feridian">https://ko-fi.com/feridian</a>

Touka (Illustrious) | Suiren no Yado (Asakawa) | Natsu no Yari Naoshi

Used 67 "clear" images for training, since there were only black and white manga pages available, and I didn't want the LoRA to be messy.

Files with SHA256: 4492e76a89393cdc8b945701efc713290b0b716bd72eb6f33363607b013c8ded