Replies: 1 comment
-
The AMD gang really should be asking AMD for better AI support instead 😄 |
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
First off, Automatic1111 is fantastic, featurewise and develops really well. Good job and keep it up!
I'd wish i'd be able to take part in tho. With a AMD GPU and windows setup its not doable to my knowledge (i know about rocm)
I want to suggest adding support with AMDs RDNA2 /vulcan api as i've seen others implement funktioning webui solutions that work very well, speed wise.
With a RX6950xt i get roughly 4.5 it/sec which is definitely acceptable.
With a ONNX conversion using ML driver i get about 1-1.1 it/sek which is livable, still about x10-20 faster than CPU only.
I know we are like a dozen people that have amd GPUs but i diden't know about stable diffusion untill after i went with AMD. hope you guys can give us AMD gang a treat.
Beta Was this translation helpful? Give feedback.
All reactions