This is Version 1. Its only been trained on about 100 images (still manually tagged) and is just a proof of concept. While I wasn't originally intending to publish this, I figured its good enough for some people to play around with while I continue to build the rest. I have tested this Lora on a few checkpoints and found good results from both semi realism and anime checkpoints but I have scratched the surface on testing. Thank you everyone, I will also release a PonyXL version once I am happy with the SDXL version.

