Deep Learning Inference Speed - Jetson Xavier versus GPU cluster


For deep learning inference speed, would the Jetson Xavier be faster than using a cluster of GPUs. I need to analyse 3468x5472 pixel images as fast as possible.

Thank you,


Inference performance depends on which model do you use.
Here are some performance report of Xavier for your reference:

Not sure which generation of your GPU cluster is.
Our suggestion is to apply the similar benchmark on your GPU cluster for comparison directly.


I would suggesting running ML-Perf inference benchmark for comparison.

1 Like