This is a slightly less baked (20% less) version of the LoRA, trained on the same dataset of images of Christina in her mid 20s, which seems to allow more flexibility and produce less saturated images while retaining fidelity.
This is a slightly less baked (20% less) version of the LoRA, trained on the same dataset of images of Christina in her mid 20s, which seems to allow more flexibility and produce less saturated images while retaining fidelity.