유형
TextualInversion
게시 날짜
2023-03-20
기본 모델
SD 1.5
트리거 단어
clrs
복사
버전 소개
Trained on 10 images with a constant learning rate of 0.05.








This embedding is going to add some color and a "splatter paint" style to your images. It is more of a creative embedding than a specific style, so who knows what you will get! I tried something a bit different and only used 10 images for training instead of the usual 25 and it came out great.
I tested it with models that do digital painting style images like Dreamshaper and Roboetics mix, but it should work with any 1.5 model.
I would love to see your images, so please share if you can make anything cool!
Trained on 10 images with a constant learning rate of 0.05.
1. 재게시된 모델의 권리는 원 제작자에게 있습니다.
2. 모델 원작자가 모델을 인증받으려면 공식 채널을 통해 SeaArt.AI 직원에게 문의하세요. 저희는 모든 창작자의 권리를 보호하기 위해 노력합니다. 인증하러 이동
