LyCORIS模型推荐在训练底模上使用，适当减少权重以获得最佳效果，通常在0.8-0.9。SDXL LoRA/LyCORIS works best on its original training model. You may need to reduce weight when using it with other models or additional LoRAs. ——by silvermoong // 如果你想要特定画师的画风，请告诉我，我会在有空的时候尝试训练。Feel free to request the artist style you want. I'll do it if I'm free. No promises on training every style. 更偏向二次元风格或有趣的画风.。以及稳定、易收集的风格更好。Mainly based on my preferences — mostly anime or interesting styles. Also, I prefer consistent styles that are convenient to collect on danbooru or pixiv. // 如果你喜欢我的模型，可以请我喝杯奶茶！https://afdian.com/a/arymodu // P.S. 最近在等CNXL的新版本，可能不会在以前的模型上训练. Lately I've been waiting for the new version of the Chenkin Noob XL model, so I probably won't be training on older models. （Noob v预测版本颜色确实很好，但肢体还是容易让人破防......）