Takedown requests can be submitted via Twitter at @AkioA_I
削除依頼はTwitter（@AkioA_I）まで提出可能です。
THIS IS ANIMA, NOT ILLUSTRIOUS OR NOOBAI! It is a different model and it won't work on those checkpoints/models!
I will be training on their latest style soon, I am just training one by one.
Description
FAQ
Comments (11)
The workflow I am using seems to not save the metadata or something. Sorry!
hi, boss, please share the link to the artist, cant find((
@mifink94 https://x.com/aruhshura
@AkioAI much thanks)
@AkioAI and thanks for lora too, excellent!
@mifink94 you're welcome 😁
@AkioAI What are your impressions of Anima in contrast to Illustrious and NAI, if itst secret?
@mifink94 Well, to me, Anima is quite better at concepts and describing what you want in the image. It supports natural language (example "a girl alone") and booru tagging (example "1girl, solo").
The only thing I dislike about it is how the styles when prompting an artist without any lora styles, it is not as strong as I want it to be. I am hoping the full model would be better than the preview when it comes to styles. It is still good at styles but for me it is not as strong with styles compared to Illustrious and NAI but It could just be me.
You can try it out yourself to see the differences.
@mifink94 But I also haven't really tried a lot of artists on it so don't go off my word as fact. Here is a website that someone made that they have tested artist styles themselves on anima preview model: https://thetacursed.github.io/Anima-Style-Explorer/
@AkioAI ouuu, thanks for such a detailed answer. I just had a negative experience with Lumina. But this model really amazed me with its clarity and understanding of the language. I guess The ILL and NAI models "hold" styles so well thanks to the many years of work of developers like you:)
I think it will be even better soon.
@AkioAI oh, this is a really cool resource, thank you again. I look forward to your new loras))