Hi :D
This is a Lora trained on Chainsawman anime and AOT both from MAPPA Studio.
It produces pics similar to those animes and it doesn't have a trigger word.
Recommanded settings:
Weight 0.2-0.6 (0.4~).
Clip skip: 2-3
I recommand using character Loras for better results.
Please keep in mind that this is a beta version so don't expect much, but I'm trying my best to learn and figure out how Lora works (cuz i'm a beginner in Lora training) and i will keep updating it as frequently as i can :D
Now available on Tensor Art :)
If you wanna support me please follow me on twitter:
You can also check out my Pixiv:
https://www.pixiv.net/en/users/90111017
Description
Beta v1.0
I tried a diffrent training method
Looking forward for your feedback :D
FAQ
Comments (4)
I request kyoani style.
I will consider that
@Headshot_btw Do you use VAE to make it look like an anime screenshoot?
@satoshipanutanku861 I tried both but i prefer when i use vae
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.