👉 Re-trained to hopefully avoid hair and eye color mix-up, with pictures of Yamato and Sora together (which I did not do in V1);
👉 Re-tagged all images and improved clothing tags;
👉 Specifying hair and eye colors not needed when generating them.








Sora Takenouchi and Yamato / Matt Ishida from Digimon Adventure tri. for Illustrious~
Trigger words are sora_takenouchi_tri and yamato_ishida_tri.
Is it necessarily better than V1? It depends, you can be the judge of that... But what's new in this version?
👉 Re-trained to hopefully avoid hair and eye color mix-up, with pictures of Yamato and Sora together (which I did not do in V1);
👉 Re-tagged all images and improved clothing tags;
👉 Specifying hair and eye colors not needed when generating them.
Tags for their uniforms:
Sora: light blue blazer, white collared shirt, red bow, grey sweater vest, pleated skirt, kneehighs (for the summer version, remove "light blue blazer" and add "short sleeves")
Yamato/Matt: light blue blazer, white collared shirt, navy necktie, belt, pants (for the summer version, remove "light blue blazer" and add "sleeves rolled up")
Ahh, a very self-indulgent LoRA... Sora Takenouchi AND Yamato Ishida from Digimon Adventure tri. in 1 LoRA for my shipping needs! :)
I made them separately at first to try my hand at this LoRA making thing, then decided having them in a single LoRA could give me some better results if I wanted them together in pictures. I don't know. This was nearly a shot in the dark (except from some help from this great article), but I think I've sorta made it.
Trigger words for them in their uniforms:
sora_takenouchi_tri, blazer, collared shirt, bow, pleated skirt
yamato_ishida_tri, blazer, collared shirt, necktie, belt, pants
For their summer uniforms, just remove "blazer" and it should give you the right outfits.
Tip: Specifying their hair and eye colors is not needed, but can help. Depending on the checkpoint, "red eyes" will give Sora's eyes a closer match to the anime than "maroon eyes". You decide what works best for you. :)
As always, constructive criticism is welcome and greatly appreciated.
👉 Re-trained to hopefully avoid hair and eye color mix-up, with pictures of Yamato and Sora together (which I did not do in V1);
👉 Re-tagged all images and improved clothing tags;
👉 Specifying hair and eye colors not needed when generating them.
1. The rights to reposted models belong to original creators.
2. Original creators should contact SeaArt.AI staff through official channels to claim their models. We are committed to protecting every creator's rights. Click to Claim
