NVIDIA has unveiled its new Jetson AGX Orin 32GB production module as it targets firms looking to bring AI and robotics applications and products to market.
According to NVIDIA, 32 technology providers in the NVIDIA Partner Network are offering commercially available products powered by the new module, which “provides up to a 6x performance leap over the previous generation”.
Via the offerings from Jetson partners, the idea is to let developers build and deploy Orin-powered systems for cameras, sensors, software and connectivity suited for edge AI, robotics, AIoT and embedded applications.
The tech company reckons its new module will appeal to industries such as manufacturing, retail, construction, agriculture, logistics, healthcare, smart cities and last-mile delivery.
NVIDIA states: “Traditionally, developers and engineers have been limited in their ability to handle multiple concurrent data streams for complex application environments. They’ve faced strict latency requirements, energy-efficiency constraints, and issues with high-bandwidth wireless connectivity. And they need to be able to easily manage over-the-air software updates. They’ve also been forced to include multiple chips in their designs to harness the compute resources needed to process diverse, ever-growing amounts of data.”
- NVIDIA Unveils Hybrid Quantum-Classical Computing Platform – read the news here
Naturally, NVIDIA reckons its Jetson AGX Orin overcomes these challenges.
The Jetson AGX Orin developer kit is capable of up to 275 trillion operations per second, and supports multiple concurrent AI application pipelines with an NVIDIA Ampere architecture GPU.
Customers could develop solutions using AI models to solve problems such as natural language understanding, 3D perception and multi-sensor fusion.
The Jetson AGX Orin 32GB module is available to purchase now, while the 64GB version will be available in November. Two Orin NX production modules are coming later this year.
The production systems are supported by the NVIDIA Jetson software stack.
On top of JetPack SDK, which includes the NVIDIA CUDA-X accelerated stack, Jetson Orin supports NVIDIA platforms and frameworks such as Isaac for robotics, DeepStream for computer vision, Riva for natural language understanding, TAO Toolkit for model development with pretrained models, and Metropolis, an application framework, set of developer tools and partner ecosystem that brings visual data and AI together.
NVIDIA points out that more than one million developers and over 6,000 companies are building commercial products on the NVIDIA Jetson edge AI and robotics computing platform.
In other news, last month NVIDIA launched a unified computing platform for quantum research and development across AI, HPC, health, finance and other disciplines.
Called NVIDIA Quantum Optimized Device Architecture, or QODA, this is the successor to CUDA – Compute Unified Device Architecture.
NVIDIA’s ambition is to solve the long-standing issue in the quantum world: to build quantum interoperability, create better accessibility for traditional DevOps roles, all while preserving its open nature.
Image courtesy of NVIDIA.