This hypernetwork was trained on AnythingV3, so it should work pretty well with any or almost any other anime-based model. Also shows pretty solid results with Protogen, so feel free to experiment.
Showcase images was generated with AnythingV3.
My prompts for experemintation was pretty simple, so i recommend to begin with this parameters:
Prompt:
masterpiece, your_object, dark_background, flat illustration, vector, very detailedAlso try
masterpiece, your_object, dark_background, very detailed, icon, centered, realistic, cinematic lightingNegative:
low quality, lowres, text, error, (cropped:1.2), worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, black and white, monochrome, censored, (out of frame:1.3)
Please share your images in review section down below.
Description
It's not exactly v2, but just another variant. Different dataset, more training steps, black outline and more iconish style than stickerish.
Sometimes gives not exact color images, so just put color which you wanna exclude in negative prompt.