Yuzus goal are easy to archive high quality images with a style that can range from anime to light semi realistic (where semi realistic is the default style). Be aware that some prompts can push it more to realism like "detailed". If faces apear more near the viewer, it also tends to go more realistic.
My Discord for everything related to anime models and art. You can support me on my and if you are interested, we also have a collaboration of multiple AI artists named that you can support!
Baked in VAE is clearvae_variant. You can also use klF8Anime2, to push the colors, but eyes can get blurry.
Supporters:
Thanks to my supporters Riyu, SETI, Jelly, lawrencium_Lr103, Alessandro and Kodokuna on my !
You can support me on my , where you can get early access to newer model versions.
_____________________________________________________
Using the model:
Use mostly tags. No extra vae needed. For better promting on it, use this or . But instead of {}, use (), stable-diffusion-webui use (). Use "masterpiece" and "best quality" in positive, "worst quality" and "low quality" in negative.
My negative ones are: (low quality, worst quality:1.4) with extra monochrome, signature, text or logo when needed.
Use clip skip 1 or 2 with sampler DPM++ 2M Karras or DDIM. I use clip 2.
Don't use face restore and underscores _, type red eyes and not red_eyes.