serverless RuntimeError: CUDA unknown error
My serverless functions keep throwing the error: RuntimeError: CUDA unknown error - this may be due to an incorrectly set up environment, e.g. changing env variable CUDA_VISIBLE_DEVICES after program start. Setting the available devices to be zero.
Upon investigation, I noticed that the failing workers all have Driver version 580.126.09, whereas those with Driver version 580.126.16 do not encounter this issue. Is there a way to force allocation to machines running Driver version 580.126.16
Upon investigation, I noticed that the failing workers all have Driver version 580.126.09, whereas those with Driver version 580.126.16 do not encounter this issue. Is there a way to force allocation to machines running Driver version 580.126.16
