Could you share me with some intuitive examples as below which tell the use cases for different TOPS value?(in the range of 0.5-10 TOPS)

Hello,

From the document (Jetson_Orin_Nano_Series_Jetson_Nano_Jetson_TX2_NX_Interface_Comparison_Migration_DA-11084-001_v1.1.pdf), we can know that Jetson Nano’s AI performance is 0.5 TFLOPS(Dense) and Jetson Orin Nano 4GB is 10 TOPS(Dense).

So, Could you share me with some intuitive examples as below which tell the use cases for different TOPS value?(in the range of 0.5-10 TOPS)

Decoding AI Performance on RTX AI PCs | NVIDIA Blog

Thanks, Steven

Hi,

You can find some benchmark data below:

To use Orin Nano with LLM use case will look like below:

Thanks.

Hi,

Thanks for your repy.
I checked Jetson Benchmarks from the link: Jetson Benchmarks | NVIDIA Developer,
but Jetson Nano and Jetson Orin Nano 4GB’s Benchmarks can not be found.

What AI application scenarios can Jetson Nano (0.5 TFLOPS) and Jetson Orin Nano 4GB(10 TFLOPS) be sufficient for? For face recognition application, how many channels and how much definition video can they achieve?

Thanks, Steven

Hi,

We stop to support Jetson Nano so there is no benchmark result for the device.
But you can find vision-based performance data of Orin Nano below:
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Performance.html#id43

Thanks.