MLPerf benchmarck on Jetson AGX Orin


I would like to benchmark my Jetson AGX Orin using MLPERF here, but I am having some issues with key rotation.

And in this tutorial you have said that if we have some issues to contact NVIDIA:

So I did and contacted NVIDIA but I get no replies or any help !!

here is the links where I explains very well my problem with photos and I contacted as well in private messages:

So how can I run this benchmarck on my Jeston AGX Orin ?


I tried and saw plenty of solutions (or hack if you want to call it) on your NVIDIA topics and on GitHub as well, like modifying the Dockerfile and adding the commands in it, etc… but nothing worked for me and most of the issues I saw was for a different architechture of mine, while mine is the Jetson AGX Orin.

Thank you in advance and I hope that I will get some help for this topics this time.

Best regard,

Hi @Harry-S, the MLPerf results on Jetson AGX Orin are a preview, so they cannot be reproduced at this time. A preview SW was used for this submission. The next MLPerf Inference submission for Jetson AGX Orin will be able to be run by other users.

1 Like

Hello @dusty_nv ,

Thank you for this clarification.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.