Either increase mnist benchmark's computation time (https://github.com/laszewsk/mlcommons/blob/main/examples/mnist-tensorflow/mnist.py) or run resnet benchmark on rivanna and use this as a benchmark (https://www.hpcwire.com/off-the-wire/resnet-50-score-confirms-leading-inference-performance-of-groq-processor/#:~:text=ResNet%2D50%20is%20an%20inference,performance%20of%20machine%20learning%20accelerators.)
Either increase mnist benchmark's computation time (https://github.com/laszewsk/mlcommons/blob/main/examples/mnist-tensorflow/mnist.py) or
run resnet benchmark on rivanna and use this as a benchmark (https://www.hpcwire.com/off-the-wire/resnet-50-score-confirms-leading-inference-performance-of-groq-processor/#:~:text=ResNet%2D50%20is%20an%20inference,performance%20of%20machine%20learning%20accelerators.)