Handcuffed-LoRA-V1 for SDXL

A LoRA model as an extension for Stable Diffusion XL (SDXL)

The LoRA allows you to easily create images of handcuffed people. It is very flexible and has been trained with the arms in front and behind the body, as well as standing and kneeling. It has a tag specifically for close-up shots of handcuffed hands and also allows easy full body shots and specific angles.

Since hands and fingers are almost always in the picture due to the theme, the typical weaknesses of this generation of AI naturally show up in many pictures. Too many fingers, thumbs on the wrong side, etc. are not uncommon. There are also images in which everything is fine, the promo images for example are completely unprocessed but of course a selection of the better results.

For perfect pictures mostly image postprocessing will be necessary.

It was trained with about 400 images and 13 tags and can be used with many styles.

The package contains 4 LoRA

Version 1.0 is the version I consider most suitable and flexible.

The other versions are less (0.9) and higher (1.1 & 1.2) trained and behave in many cases accordingly different. In certain situations these might be the better choice, which is why they are included in the package.

Of course you need Stable Diffusion XL and one of the many user interfaces to use it. A small description is included.

All product images are unedited and were created using the SDXL base model and the Fooocus user interface.

In addition, the readme file contains 4 example prompts, which were also used to create the promo images, these are intended as a starting point.

the password for the zip files is always "Frotz"