Chibizilla

I trained the LoRA with 100 images of varying amounts of silliness and cuteness in all kinds of poses, colors and various backgrounds (most of them very simple). So it should be very flexible when specified further by the user.

Recommended tags: 1girl, chibi, solo, godzilla (cosplay), tail

First tests work really well. Have fun experimenting. I'm looking forward to everyone's results. P.S.: Atomic breath images are lucky hits at best in my experience. Eventually I would like to create a LoRA that allows for any kind of characters to shoot energybeams from their mouths, but as of yet the A.I. stubbornly refuses to do so.