这是一个来自B站的不知名虚拟主播的Lora(可以自行猜测)
This is Lora, an unknown virtual anchor from station B (you can guess by yourself)
通过大量混合,它的对原人物的特点还原能力已经下降了很多,但也因此获得了更好的泛化能力
Due to extensive mixing, the model's ability to faithfully restore the original character's traits has significantly decreased, but it has gained better generalization capabilities.
该模型的特点是:简单背景和一个不错的打光
The model's features include simple backgrounds and good lighting
角色本身拥有两套服装,可以分别通过关键词“school uniform”和“lolita"触发
The character has two outfits that can be triggered separately using the keywords "school uniform" and "lolita"
在两个词都不存在时,则会随机触发其中的一件
If both keywords are absent, the model will randomly select one of the outfits
推荐的权重是0.5-0.8之间,可以自行尝试, 不建议超过1.0
The recommended weights for the keywords are between 0.5 and 0.8, but it is advised not to exceed 1.0
推荐使用的大模型是Counterfeit2.5,当然你也可以自己尝试
The recommended large model is Counterfeit2.5, of course you can also try it yourself
为了这个模型,我花了很多心思,希望你能用的愉快
I put a lot of effort into creating this model, and I hope you enjoy using it. Have fun!