The model I think is fairly self explanatory.

Additional 0.5 version appended, works best when used at the same time as the prototype, doesn't seem to work very well on it's own and does more of a 'clasped footjob' on it's own.

This ones a bit unreliable, limited dataset that will likely have to be expanded. I have an idea to generate enough images using this LoRA to generate enough images that -work- and coming with a 2.0 version later on, this version currently here is a placeholder and more of a prototype. Previews made with novaFurryXL_illustriousV20.