NVIDIA has launched the Jetson Orin Nano 2, a new robotics computer delivering twice the AI inference performance of its predecessor within the same compact form factor. The system achieves this leap by utilizing improved Tensor Cores and higher memory bandwidth, while also reducing power consumption by 40% at equivalent performance levels, NVIDIA says.
More than 3 million developers already utilize the NVIDIA robotics stack, and companies like Cognex, Doosan Bobcat and Matic are early adopters of the new technology. “Today’s small and medium models have reached the accuracy of last year’s largest models,” said Deepu Talla, vice president of robotics and edge AI at NVIDIA, “unlocking real-time intelligence for edge devices.”
Jetson Orin Nano 2 Delivers 2x Inference, 40% Less Power
This increase in capability allows for more complex AI tasks to be processed directly on the device, reducing reliance on cloud connectivity and lowering latency for applications like robotics and drone navigation. The new module delivers this performance increase while consuming 40% less power in 15-watt mode compared to its predecessor, extending battery life for mobile and remote deployments, according to NVIDIA.
NVIDIA reports the Jetson Orin Nano 2 features 78 trillion operations per second of AI compute alongside 8GB of memory and an 8-core Arm CPU, positioning it as a cost-effective solution for edge AI applications. Wing, a drone delivery company, intends to evaluate the Jetson Orin Nano 2 to potentially improve the responsiveness and energy efficiency of its drone fleet. Dinuka Abeywardena, head of perception at Wing, explained that drone delivery depends on AI that can enable fast, reliable understanding of the real world.
Matic Robotics is also adopting the new module to enhance its home cleaning robots, enabling features like conversational AI and precise mapping. Navneet Dalal, cofounder and CEO of Matic Robotics, explained that with Jetson Orin Nano 2, Matic can run AI models at the edge in a compact home robotics platform built for real-time perception, interaction and navigation.
Today’s small and medium frontier models have reached the accuracy of last year’s largest frontier models, unlocking real-time intelligence for edge devices.
Deepu Talla, vice president of robotics and edge AI at NVIDIA
See today’s quantum computing news on Quantum Zeitgeist for the latest breakthroughs in qubits, hardware, algorithms, and industry deals.
