RTX 3090 : cudaMallocAsync

[metamountain]

Cuda not accepted. Device: cuda:0

Loading Log:

Set vram state to: HIGH_VRAM
Device: cuda:0 NVIDIA GeForce RTX 3090 : cudaMallocAsync
VAE dtype: torch.bfloat16
Using pytorch cross attention

Cuda not accepted. Device: cuda:0

How can I solve this problem ?
I guess acceleration is not working correctly?

[wangsan0315]

Same problem, can't find a solution

[kurt83340]

im not pro and maybe i will say a crap but i think everythink is ok and just
your graphic card is on the port 0 .... this number can be one if you have multiple graphic card

so everything is ok .... according to me ... but im not programmer and all people here are smarter than me

[specblades]

Same here