NVIDIA has set new records in AI inference and per-accelerator performance at the edge with its on-chip system based on NVIDIA Ampere architecture, the NVIDIA Orin.
Debuting its MLPerf benchmarks, the Orin accelerator showcased the highest performance and largest base for running all machine-learning workloads and scenarios.
In terms of edge AI, a pre-production version of the NVIDIA Orin came out on top in five of the six performance tests. It ran up to five times faster than the company’s previous model, the Jetson AGX Xavier, whilst maintaining an average energy usage twice as efficient.
NVIDIA Orin is currently available in the NVIDIA Jetson AGX Orin developer kit for robotics and autonomous systems. The Jetson platform is used for AI inference and other tasks by clients such as AWS, John Deere, and Microsoft Azure.
The Orin dev kit is also a key component of NVIDIA’s Hyperion platform for autonomous vehicles. BYD, China’s largest electric vehicle manufacturer, will use the Orin-based DRIVE Hyperion architecture for its next-generation automated electric vehicles.
In what was the fifth round of MLPerf benchmark testing for production AI, devices with NVIDIA GPUs running the Jetson AGX Orin were the only edge accelerators to run all six benchmarks.
Even where other devices could run certain benchmarks, NVIDIA showed leading performance across the board.
The MLPerf benchmarks are widely supported by the industry, with recognition from companies like Amazon, Arm, Facebook, Google, Intel, and Microsoft.
NVIDIA unveiled the Jetson AGX Orin at its GTC event in November last year.
At the time, NVIDIA’s VP and general manager of embedded and edge computing, Deepu Talla, said: “As robotics and embedded computing transform manufacturing, healthcare, retail, transportation, smart cities, and other essential sectors of the economy, the demand for processing continues to surge.
“Jetson AGX Orin addresses this need, enabling the 850,000 Jetson developers and over 6,000 companies building commercial products on it to create and deploy autonomous machines and edge AI applications that once seemed impossible.”
Want to learn more about edge computing from industry leaders? Check out Edge Computing Expo taking place in Amsterdam, California and London.
Explore other upcoming enterprise technology events and webinars powered by TechForge here.