Trained with many images of in game selfies and screen shots, for all you VR nerds but we all know beat saber is better

The models I make are tested for resolutions under 512 (Not by choice), results may be different (Probably better) with higher resolutions. Ref images are in the lower quality as well.

Honestly, donate a penny and I'd be happy
Ko-fi: https://ko-fi.com/keyennorman

Prerequisites, straight to the point, quite a bit to go over

1. Wights fluctuate quite a bit. I've tried from 1 to 0.5 etc and it seems like all of them COULD work under the right circumstance BUT I'd recommend starting with 0.7-1
2. This LORA is large for a reason, wanted to make sure the style was correct and fool proof, maybe in the future I can update the model to a more optimized state but for now you better have a spare USB or memory card laying around somewhere, loser
3. As for other LORA capability there is potential, I encourage you to experiment to see which LORAs work the best with this model.
4. I find the style struggles with male bodies as well as basic actions such as HOLDING. But that's to be expected in all honesty. Maybe it was just me who had the problem.
5. Probably can generate futa images, for you cultured folk. I haven't tried it myself.

Hopefully your experience with this model proves fruitful in some way, cheers.