Please contact me here on civitAi or on Twitter if you want to provide the LoRa in a generation service.

TLDR (Because I don't care)

Use cinderella <lora:cinderella_v10:1> in positiv promt.

If you like the LoRa, please feel free to share your generated images here and leave a review (or at least a rating) ;D

About

I don't know why, but I felt the need to create a Cinderella LoRa, I don't really like the story (especially from the second film, that's no story...), so I think I just did it because Cinderella is cute... Overall I would consider the model a failure, didn't work out as expected and only in some cases the model produces what I expected...

Training

I gathered 47 images from both movies, but the majority came from the first movie (I think the second movie was one of these VHS only releases so not so good images and the art style was a bit different but okay... I was able to work with it. This was my first try using tags to define different outfits, and I screwed up a bit with her "normal" outfit, see the usage section below. Eventhou it was a failure I learned quite a lot ^^

Usage

No weighing is required, so it's as simple as that:

cinderella <lora:cinderella_v10:1>

This is the first LoRa where I tried to train the different outfits of the character... so there are 4 outfits with corresponding trigger words:

medieval maid - casual/work outfit if your checkpoint knows the concept of maids (aka any anime checkpoint), add (maid:2) into your negative promt, couldn't think a better name for this outfit ...

princess dress - princess dress

ripped dress - when her original dress was ripped by her stepsisters

nightgown - from the first scene of the first movie (was only trained with 2-3 images, so it is a bit hard to get)

I think the best result I got was with dreamshaper, with any other model she didn't looked quite right (but still cute ;D)