Trained on AnyLora
To get the best results, please read the description below:
Recommended weight: 0.7-0.9
Activation word: Rebecca
Upscale to 1.5
Using anime SD models is recommended, semi-realistic checkpoints should also work
I would appreciate any reviews, positive or negative!
I would very much appreciate you sharing your results ^_^
If you have any suggestions on what LoRA i should do next, please share them in the comments
Enjoy :)
Description
Version 1 - trained on 58 images
FAQ
Comments (4)
I assume this isn't including her armor?
Yes, although i plan to update this LoRA with a better dataset which would include her armor.
The realistic texture and coquettishness of the characters' skin in this batch is incredible. I have never seen a batch that can produce this kind of expression. However, I can't get the same texture using 12user34kn276's other One Piece characters' loras (it's difficult to reproduce this kind of expression using prompts and add-details lora). It would be great if other characters could have the same texture as Rebecca's lora.
Thank you for the kind words!
Unfortunately, i do not yet possess sufficient skills to freely manipulate texture or style of my character LoRA’s. The end results are a mystery until i start testing the LoRA’s and the resulting texture or style mainly depends on the dataset - and that is something i have very little control over. I rarely have surplus images for training as i pick the best from various sources on the internet.
Best i can do right now is filter those images to try and achieve a particular style but this very rarely works out :(
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.