Very much a WIP. I'm still learning about making loras.This model is over-trained. It'll only give the same few images over and over again unless you turn the weight down a lot. .4-.6 tends to work. I'm really just putting it up here as an easy way to share with a few people, so sorry for the lack of details.It's only really gonna work with women and side views.

Inflated Airbag

This version includes different angles. The previous one only really worked with side views. It doesn't do well with txt2img, but img2img works great. This was made using the CivitAI trainer.

Files with SHA256: 4e08fda7ed39eb47eca12d584e7c30f3862ea9e8be18f34ad7668b5420134606