LoRA for the Tyler Sisters from Yu-Gi-Oh! Arc-V.
Surprised nobody uploaded this yet, so went ahead and made one. It turned out well I think. I trained it with 195 pictures in which in which about 175 were screencaps I manually took and edited and the rest was a few fan drawings. I took and edited it at 20 epochs with a dim/alpha of 64/32.
Triggers:
To trigger the character "Grace Tyler" just input the following after the LoRA : Grace Tyler, white hair, yellow eyes, 1girl, solo
To trigger the character "Gloria Tyler" just input the following after the LoRA : Gloria Tyler, blonde hair, pink eyes, 1girl, solo
To trigger both girls in one picture just input both triggers above except replace "1girl, solo" with "2girls, multiple girls".
Notes:
My experience has shown scale 0.5 ~ 0.95 work best; my best results were 0.65.
My prompt order shown in the sample pictures worked best for me.
Works on different models though weights may need to change for good results. Refer to the last two pictures in which the NeverEnding Dream (NED) and MeinaMix models were used.
I used FoxyaNSFW model for most of the pictures and it could give me a wide variety of styles.
Refer to my pictures for examples and what upscaler I used, keep in mind I have used other textual inversions to help with poses and negatives.
Even though the training data was nearly 95% safe for work, it can give nsfw without forcing it. If you want to use this for sfw, remember to put nsfw in negative prompts etc.
Works very well when creating a picture of each girl, but does struggle when creating a picture with both in the same frame. Their eye colours often get switched for some reason. Easy solution is to inpaint over the wrong coloured eyes and simply prompt for the eye colours you want.
~ ~ Spunk yes ;) ~ ~