More than 30 leading technology partners worldwide announced this week the first wave of production NVIDIA Jetson AGX Orin-powered systems at COMPUTEX in Taipei.
The new products come from dozens of camera, sensor and hardware providers in Taiwan for use in high-end AI, AI, robotics and embedded applications.
Available worldwide since GTC in March, and Nvidia Jetson AGX Oren The developer suite delivers 275 trillion operations per second, packing over 8 times the processing power of its predecessor, the NVIDIA AGX Xavier, into the same pin-compatible form factor.
The Jetson Orin features NVIDIA Ampere GPUs, Arm Cortex-A78AE CPUs, next-generation vision and deep learning accelerators, high-speed interfaces, faster memory bandwidth, and multimedia sensor support capable of feeding multiple simultaneous pipelines for AI applications.
Delivering server-class performance for advanced AI, the new Jetson AGX Orin production units will be available in July, the Jetson AGX Orin production units will be available in July, and the Orin NX units will be available in September.
“The new Jetson AGX Orin ships the next generation of robotics and cutting-edge AI applications,” said Deepu Tala, Vice President of Embedded Computing and Computing at NVIDIA. “This momentum continues to accelerate as our ecosystem partners release Jetson Orin-based production systems in various formats designed to suit specific industries and use cases.”
Strong product and partner ecosystem
Jetson-based products announced include servers, high-end hardware, industrial computers, bus boards, artificial intelligence software, and more. It will be available in fan and fanless configurations with multiple connectivity and interface options, including also specifications for commercial or robust applications in robotics, manufacturing, retail, transportation, smart cities, healthcare, and other essential sectors of the economy.
Among these versions are members of NVIDIA Partner Networkincluding AAEON, Adlink, Advantech, Aetina, AIMobile, Appropho, AverMedia, Axiomtek, EverFocus, Neousys, Onyx and Vecow.
Other NVIDIA partners that have launched new Jetson Orin-based solutions worldwide include Auvidea, Basler AG, Connect Tech, D3 Engineering, Diamond Systems, e-Con Systems, Forecr, Framos, Infineon, Leetop, Leopard Imaging, MiiVii and Quectel, RidgeRun, Sequitur, Silex, SmartCow, Stereolabs, Syslogic, Realtimes, Telit, TZTEK, to name a few.
Million Plus for Jetson Developers
Today, more than 1 million developers and more than 6,000 companies build commercial products on the NVIDIA Jetson edge AI computing platform and robotics to create and deploy autonomous machines and cutting-edge AI applications.
And with more than 150 members, the growing number Jetson ecosystem Partners offer a wide range of support, including from companies specializing in AI software, hardware and application design services, cameras, sensors and peripherals, developer tools and development systems. This year, the AAEON BOXER-8240 powered by Jetson AGX Xavier won the COMPUTEX 2022 Best Gold Choice Award.
Developers build their next-generation applications on the Jetson AGX Orin development kit for seamless deployment on production units. Jetson AGX Orin users can benefit from a file NVIDIA CUDA-X accelerated computing stack, nvidia jetpack SDK and the latest NVIDIA tools for application development and optimization, including cloud-native development workflows.
Comprehensive software support
Jetson Orin enables developers to deploy the largest and most complex models needed to solve the challenges of advanced AI and robotics in natural language understanding, 3D perception, multi-sensor integration, and other areas.
“NVIDIA is the recognized leader in artificial intelligence and continues to leverage this expertise to develop robotics through a robust ecosystem and end-to-end solutions, including a suite of hardware platforms that leverage popular tools and neural network models,” said Jim McGregor. Principal Analyst at TIRIAS Research.
“The new Jetson platform delivers the performance and versatility of the NVIDIA Ampere architecture to enable further advances in autonomous mobile robotics for a wide range of applications ranging from agriculture and manufacturing to healthcare and smart cities,” he said.
Pre-trained models from NVIDIA NGC The catalog is optimized and ready for fine-tuning with NVIDIA TAO Toolkit and customer data sets. This reduces time and cost for production-quality AI deployments, while cloud-native technologies allow for seamless updates throughout the life of the product.
For specific use cases, NVIDIA software platforms include files Nvidia Isaac Sim On the Omniverse for Robots; Riva, GPU Acceleration SDK for Building Speech AI Applications; The deep stream Stream Analytics Toolkit for AI-based multi-sensor processing and video, audio, and image understanding; Beside Metropolisan application framework, a suite of developer tools and a partner ecosystem that brings visual data and artificial intelligence together to improve operational efficiency and safety across industries.
Watch NVIDIA’s Computex main address Monday, May 23 at 8 p.m. PT.