The Connect Tech Boson carrier board was used with the new NVIDIA® Jetson Orin™ NX module for an MLPerf™ Inference v3.0 submission. The results showed up to a 3.2X inference speedup compared to the previous-generation Jetson Xavier™ NX.
Customers are not limited to the Boson carrier board to enjoy these performance gains. As an Elite partner in the NVIDIA Partner Network, Connect Tech additionally supports NVIDIA Jetson Orin NX with Photon and Hadron carrier boards, and edge devices like Polaris and Rudi-NX. By deploying on Connect Tech’s production-ready hardware, customers can take immediate advantage of Jetson Orin NX for performance improvements and enhanced user experience with robotics and other edge AI applications.
What is MLPerf™ Benchmarking?
When it comes to AI, benchmarking is the preferred tool to identify the best-performing technologies in the industry. Benchmarking compares measured performance across a range of AI hardware, software and services. Specialists from academia, research and industry gather and assess submissions from the AI community and provide objective results. To stay up to date, MLPerf™ continuously develops, adding new tasks and workloads alongside the evolution of AI. MLPerf™ allows a meaningful and fair comparison between AI systems.
Read more about MLPerf™ here: https://mlcommons.org/en/
Learn more about the NVIDIA submission for MLPerf Inference 3.0.
Watch this video showcasing performance gains between Xavier NX and Orin NX: