昨天那个初音的帖子好像烂了,私信也发不了,我把tag放你这下面吧

正面:((cyberpunk mask)),mouth mask,(((extremely detailed CG unity 8k wallpaper))),(((masterpiece))),(((best quality))),(((1girl))),((beautiful detailed eyes)),(((extreme detailed illustration))),cinematic lighting,art of light novel,highres,original,extremely detailed walpaper,shiny hair,((detailed face)),(mecha clothes),mecha bodysuit,(filigree),((upper body)),((((miku)))),latex stockings,mature female,blue hair,light smile,light blush,twintails,in the cyberpunk city
负面:lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits,cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, bad feet,lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry,lori,2girls,long neck
其他参数
Steps: 60, Sampler: Euler, CFG scale: 6, Seed: 594221705, Size: 1024x640, Model hash: e6e8e1fc
备注:出图还是很稳定的,如果想要面罩的话可以在正面tag加上((cyberpunk mask)),mouth mask,只是出图不那么稳定,而且不加也可以出面罩图,只是概率低点。我个人没跑过full body,有兴趣的可以自己试试。
想要别的视角可以自己加上from behind,profile之类的tag,不过要加多几个括号加权重,不然大概率还是正面图,然后不推荐跑背面图,要跑的话最好加上looking background,不然容易烂图,因为ai好像不是很会画初音的后脑勺
