🤝如您有模型定制或其他合作、咨询需求，请联系Wechat: 13439222668。请简单注明来意，时间有限不闲聊哈，谢谢理解！If you have any model customization or other collaboration and consultation needs, please contact Wechat: 13439222668. Kindly provide a brief description of your inquiry, as there's limited time and no time for casual chats. Thank you!这是一个人物与影调功能兼具的模型。我从<a target="_blank" rel="ugc" href="https://civarchive.com/models/33208/filmgirl-film-grain-loha-and-lora">FilmGirl 模型</a>的训练集中优选了颜值非常出众的部分照片训练得到了这个模型。总体而言，这个模型有与<a target="_blank" rel="ugc" href="https://civarchive.com/models/33208/filmgirl-film-grain-loha-and-lora">FilmGirl模型</a>类似的影调。相比<a target="_blank" rel="ugc" href="https://civarchive.com/models/33208/filmgirl-film-grain-loha-and-lora">FilmGirl模型</a>，“颜”模型的人像颜值更高、图像畸形率更低、但图像真实感稍弱。（毕竟FilmGirl模型的训练集与正则化图像数量要多很多）请注意：<ol><li>我常用的CFG范围是6~7，Lora权重通常在0.4~0.7。</li><li>Clip skip设为1或2时，生成的效果会有所不同，大家可以多尝试看自己喜欢哪种风格。</li><li>本LoRA模型的训练集主要为亚洲胶片人像照片，生成欧美人像的效果会受影响。</li></ol>This is a model that combines both character and tonality features. I have carefully selected a portion of highly attractive photos from the <a target="_blank" rel="ugc" href="https://civarchive.com/models/33208/filmgirl-film-grain-loha-and-lora">FilmGirl model</a>'s training set to train this model.Overall, this model has a similar tonality to the <a target="_blank" rel="ugc" href="https://civarchive.com/models/33208/filmgirl-film-grain-loha-and-lora">FilmGirl model</a>. Compared to the <a target="_blank" rel="ugc" href="https://civarchive.com/models/33208/filmgirl-film-grain-loha-and-lora">FilmGirl model</a>, the portrait attractiveness of the "Cutie" model is higher, the image distortion rate is lower, but the image realism is slightly weaker. (After all, the FilmGirl model's training set and the number of regularization images are much larger.)Please note:<ol><li>My commonly used CFG range is 6 to 7, and the Lora weight is usually 0.3 to 0.6.</li><li>When the Clip skip is set to 1 or 2, the generated effects will be different. You can try more to see which style you like.</li><li>The training set of this LoHA model is mainly Asian film portrait photos, and the effect of generating European and American portraits will be affected.</li></ol>

LEOSAM's 颜 Cutie LoRA

采用了新的打标方法，重新进行了训练，图像畸形率相比v1明显降低Using a new tagging method and re-training, the image distortion rate is significantly lower than that of v1

Files with SHA256: 8266ac98f85d4c89606d3d544308d54be9ee462df37fa876aed4731dd8e3b173