A lora of the Dare To Dream Lingerie Lace Teddy. This was trained with only 4 images. The resulting model is slightly overbaked and without a strong prompt the training data will emerge, but I've had some decent results. I might make an updated version.

With this lora you will need to have a strong prompt. You prompt will need to include the description of the subject, pose, and environment. Otherwise, your models will look a little thick. Additionally, i recommend a strength between .5-.8. But you can also get some ok results at 1.