What if I've tried to increase the prompt adherence in Illustrious v0.1 and its siblings? It's only a test though, because I'm not entirely sure if I did it correctly. :)
Left image - original checkpoint, right image - modified checkpoint. It seems the 'characteristics' of the original remain in the modified model, but prompt adherence is better for the most part.
Another observation - VAE inclusion gets 'reset', the resulting checkpoint doesn't have it.
The great merge, ANGRA Illustrious checkpoint, is also a very nice field for experimenting! It already has nice CLIP capable of understanding complex stuff, and this try of mine just changes the outcome a bit, especially with the nasty derelict bedroom. :)
This model is really good on his own, and the fact that it can simply try fitting multiple objects into portrait image is noteworthy!
1.転載モデルは学習・共有目的のみで使用し、著作権は原作者に帰属します
2.モデルの認証は公式チャンネルでご連絡ください。クリエイターの権利保護に努めています クリックして認証
