It's an pony model, i.e works best with pony and pony's merge (this especially is blends well with different styles)
Wt- 0.8-1.0
Description
This is a better trained version, with low memory usage - dim8fp4, - approx 50mb LORA. I got good results with it. Give it a try.
Trained with around 2000 steps in kohya, pony diffuser.
Finally getting a hang of LORA training, Enjoy. Let me know if you have any request.