底模:SD1.5
主页:https://www.liblib.art/modelinfo/c9987fbebd6b4505acba844200f8d653
版本:V 2.0
1girl, barcode, belt, black_nails, blue_eyes, blue_nails, bracelet, cover, cover_page, earrings, english_text, fake_cover, fingernails, jewelry, long_hair, long_sleeves, looking_at_viewer, magazine_cover, nail_polish, pants, solo, spiked_bracelet, watch, wristband, wristwatch,(claw pose:1.3), white hair cover eyes,sitting,(((pubic))), ((((pubic_hair))))sketch, duplicate, ugly, huge eyes, text, logo, monochrome, worst face, (bad and mutated hands:1.3), (worst quality:2.0), (low quality:2.0), (blurry:2.0), horror, geometry, bad_prompt, multiple limbs, bad anatomy, crown braid, ((2girl)), (deformed fingers:1.2), (long fingers:1.2),succubus wings,horn,succubus horn,succubus hairstyle, (bad-artist-anime), bad-artist, |
这款模型是对个人风格的极致体现,中国AI绘画社区从零开始到现在,越来越多的模型开始追求特效的堆叠,那种凌乱的画面一眼看上去很厉害,细看又漏洞百出。所以我想反其道而行之,让画面更贴近人类画师,少一点细节,多一些人味。
模型特点:整体画面极其干净,彻底舍弃了所谓2D图片的“立体感”,二次元就应该平铺在纸上!
重要提示:请使用animevae或clearvae!必须使用Hiresfix!不要使用EasyNegative!
参数建议:DPM SDE K 算法,20-35步。latent采样15步, 重绘幅度0.5-0.57.
在这里推荐一个通用负面tag串:
(worst quality, low quality:2),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,lowres,worst quality,low quality,normal quality,duplicate,grossproportions,deformed,out of frame,60s,70s,80s,90s,00s,
天空之境 NullStyle V2.0迭代完成!!!通过重新炼制其父本,使其在画面干净程度,人体,手部,头像,胸像,立绘,画风,动物,场景构图等方面都有着极其优秀的表现,并且修复了1.0版本在latent 0.57以下会变糊的问题。我愿称之为二次元万能模型