Trained to produce asymmetrial pants, leggings, bodysuits, jumpsuits, etc. that have one leg removed.

This model has been in the works for a while but is still in what I would consider alpha phase. Despite the concept being quite simple, I just can't get consistent outputs (<50% success rate) even when I am using very descriptive prompts and/or an overtrained model. I think it is a dataset issue, it was quite difficult to find good quality and diversity of images. If you get good results, please post them so I can use them to train future versions. The more variation and diversity in styles, people, etc. the better future versions can be.

My workflow is compatible with the onsite generator, but due to the inconsistency of achieving the desired effect, I don't recommend using it onsite unless you have a lot of buzz to spare for experimentation.