This is AIO file (Diffusion Model+TextEncoder+VAE) for convenience. Enjoy PonyV7!
Original Resources are below:
Notice:
2025/1/3
Scaled fp8 Diffusion Model + fp16 TextEncoder + fp16 VAE is now available, have fun!
2025/1/2
I finally found the origin of the problem. In fp8 quantization of PonyV7, we need "scaled fp8". Otherwise, it'll break the critical part of the model weight.
2025/12/31
Currently, fp16 AIO has no saturation problem. And, I'll try to recreate correct fp8 AIO.
2025/12/31
I found image saturation problem in fp8 AIO, so I'll upload fixed version for that.
Description
Currently, fp16 All In One has no saturation problem.
FAQ
Comments (8)
Does one use Pony 7?
I am, at least ;-)
can this model do anything yet?
v7.0_fp16(AIO) in this page will works fine. But original PonyV7 itself is still unstable, so this AIO file is unstable, too. I hope that the PonyV7.1 will be nice.
I'd use https://github.com/silveroxides/convert_to_quant instead of trying scaled
Thank you! I'll try it!
@JohnDoeSmith Could you share the script to make fp8 scaled?
@qek O.K. I'll upload it to GitHub, later.