V2.0已更新！！！
我愿称它为二次元万能模型!彻底舍弃了所谓2 D图片的“立体感”画面干净程度极好,在人体,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,任何风格的二次元场景都可以完美兼容!
一.修复了1.0版本在latent 0.57以下画面变糊的问题。
二.通过重新炼制其父本，基本解决肢体崩坏问题。
三.画风更加洁净，削弱了艳丽程度。
四.prompt识别率提升，junk data为0，clip无偏移。
tips：VAE推荐animevae或clearvae。
————————————————————————————————————————
模型名:苍玄/ NullStyle
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！
提示:请使用animevae或clearvae。
此模型禁止任何形式的商用
Model name: Cangxuan / NullStyle
Model features: The overall picture is extremely clean, the main feature is 2D flat painting style, completely get rid of the mainstream orangemix, pastelmix, and its derivative "an AI smell, skin redness, color block accumulation" and other shortcomings, completely abandon the so-called "three-dimensional sense" of 2D pictures, the second dimension should be flat on the paper!
I found a new AI tool Shakker, a best image to image tool. You can try it via https://www.shakker.ai ,it can help you:
-Remix: Upload a picture. Just switch the prompts, and you can create stunning images in the same style.
-Style Transfer: Shakker not only extracts the style,but also switches among various styles.
Besides, Shakker also offers Object Control,Composition Control,Collage Redrawing etc.
Description
我愿称它为二次元万能模型!彻底舍弃了所谓2 D图片的“立体感”画面干净程度极好,在人体,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,任何风格的二次元场景都可以完美兼容!
一.修复了1.0版本在latent 0.57以下画面变糊的问题。
二.通过重新炼制其父本，基本解决肢体崩坏问题。
三.画风更加洁净，削弱了艳丽程度。
四.prompt识别率提升，junk data为0，clip无偏移。
FAQ
Comments (8)
pruned version plz
shishi
A very distinctive model, but it has some serious interference with Lora, but the image performance is good. Let's give it 5 stars instead
请问一下，提供的Vae是在哪里下载呢？
C站里有，抱脸站也有，右边下载按钮下面files也有
animevae我猜是NovelAI用的那个，clearvae应该是这个https://civitai.com/models/22354
建议放点原图啊，不知道你这个模型适合的参数啊
肢体很容易竹竿化，多样性不太够
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.