"Please read this before using, thank you."
Generate simple anime style portraits for SD3.5L.
(Currently Works for Simple Prompts Only!)
This is an attempt to port Anitrait FX (a Flux based model) to SD3.5L. What's the difference between this and Anitrait 35L? This one has different dataset and is trained differently. It is supposedly better at portraits at the cost of background image compared to 35L. If you want to generate sceneries, check out Anitrait 35L.
Still in Alpha Phase. Instabilities May Occur!
Use settings:
> Put trigger words at the beginning or the end of prompt.
> Weight from 0.6 to 1. Sweet spot is at 0.8 for me but start at 0.7.
> CFG I usually use 4 but try using 5 for generating text.
> Use Euler or DPM++ 2M SGM Uniform. Prefer to use Euler since the Lora is trained with Euler. I never tested others, but if you're comfortable with certain samplers, you can definitely try with that.
> I do recommend using 25 or more steps, but if you know how to SD3.5L correctly, you can do your own way.
> You can't use upscaler! It will turn your images into colorful porridge. But if you know the dark magic to get it work for SD3.5L, then please try it.
> Clip skip... I don't really know. I usually use 1, but 2 also works.
> Recommended for portrait images. I'm not talking about image orientation, but like, that portraits. I hope you understand this. But if you want to create something else then feel free to do it since I'm not really a good prompter and can't test it with good prompt. Also, SD3.5L seems good for portraits in general.
Thank you for using this model for your image generation.