A tribute to the fantastic beauty again.Please refer to the previously released Mercedez <a rel="ugc" href="https://civarchive.com/models/140943/mercedez">LoRA</a>.Lycoris is trained on the completely different dataset. So it is highly recommended to use both LoRA and LyCORIS together keeping sum weight of two models ~0.8...1.1 to get some outstanding results according to the likeness.Also please refer to the useful <a rel="ugc" href="https://civarchive.com/articles/1721/improving-results-by-using-multiple-models-of-the-same-concept-turning-it-to-11">article</a> giving a hint about mixing models to reach better quality. Many thanks to <a rel="ugc" href="https://civarchive.com/user/malcolmrey/models">malcolmrey The Great </a>for that.

Files with SHA256: 65d2f5306062f6003f52ba963bb84b7f4f99d87dd4550bef0ec07543ab96bd3a