使用来自danbooru的约30张图片的训练集所训练的一个LORA。来自少女前线的OTs12 紫衫树，只包含了原造型。使用的训练模型为MeteoriteDiffusion（<a target="_blank" rel="ugc" href="https://civarchive.com/models/26091?modelVersionId=31243）。">https://civarchive.com/models/26091?modelVersionId=31243）。</a> 如果你没用过，马上去用。非常好动漫模型，爱来自中国预期也可以在NAI、anything、abyssorange等动漫模型上生效，但是可能有一些模糊。无需特别的提示词，但可以使用1girl、beret来辅助生成。推荐权重0.5-1，或者你自己看着办————————DeepL提供的翻译————————A LORA trained using a training set of about 30 images from danbooru. from Girls frontline's OTs12 Tiss, containing only the original modeling.The training model used is MeteoriteDiffusion (<a target="_blank" rel="ugc" href="https://civarchive.com/models/26091?modelVersionId=31243">https://civarchive.com/models/26091?modelVersionId=31243</a>). If you haven't used it, go use it right away. Very good anime model, love from ChinaExpected to work on NAI, anything, abyssorange and other anime models as well, but there may be some ambiguity.No special cue words are needed, but you can use :[1girl, beret] to assist in generation.Recommendation weight 0.5-1, or you can do it yourself

Files with SHA256: 90180491f03b76d173d4b28257ab65946ca30fbe0e3c9b61b94b4ecb17519d1a