NVIDIA Unveils Jetson Orin Nano 2 With 2x AI Performance for Edge Robotics

NVIDIA has announced Jetson Orin Nano 2, a new compact robotics computer designed to bring powerful generative AI and real-time reasoning to entry-level edge devices.

The company says Jetson Orin Nano 2 delivers twice the inference performance of its predecessor while maintaining the same compact form factor. At the same performance level, it can also use 40% less power in 15-watt mode, potentially making advanced AI more practical for robots, drones and other battery-powered machines.

The new platform is aimed at developers building physical AI applications, including autonomous robots, delivery drones, inspection systems and vision-based AI devices.

NVIDIA Jetson Orin Nano 2 Targets Edge AI

According to NVIDIA, Jetson Orin Nano 2 provides up to 78 trillion operations per second (TOPS) of AI compute, along with 8GB of memory and an 8-core Arm CPU.

The company says improvements to its Tensor Cores and memory bandwidth allow the new computer to deliver twice the inference performance of the Jetson Orin Nano Super while retaining the same compact form factor.

Power efficiency is another major focus.

In 15-watt mode, NVIDIA says Jetson Orin Nano 2 can deliver the same performance as its predecessor while consuming 40% less power.

That combination could be particularly important for machines operating away from traditional power sources, including drones and mobile robots.

Smaller AI Models Are Moving Intelligence to the Edge

NVIDIA is positioning the Jetson Orin Nano 2 around the growing ability of smaller AI models to perform sophisticated tasks locally.

The company says today’s smaller and medium-sized frontier models can achieve accuracy comparable to some of the largest frontier models from the previous year. That could allow more AI processing to happen directly on physical devices instead of relying entirely on cloud computing.

“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,” NVIDIA Vice President of Robotics and Edge AI Deepu Talla said.

He added that Jetson Orin Nano 2 is designed to give millions of developers access to the performance and energy efficiency required for real-time reasoning at the edge.

Jetson Orin Nano 2 Supports Generative AI Models

The new platform is powered by NVIDIA’s software ecosystem and is designed to run memory-efficient AI models optimized for edge inference.

NVIDIA says developers can use the platform with large language models and vision-language models, including NVIDIA Cosmos and Nemotron, as well as open models such as Gemma 4 and Qwen 3.

The goal is to let developers build systems that can understand images, language and their surroundings while responding in real time.

That could expand the role of AI beyond traditional software applications and into physical machines capable of sensing and acting in the real world.

More Than 3 Million Developers Use NVIDIA’s Robotics Stack

NVIDIA says more than 3 million developers have built on its robotics stack.

Jetson Orin Nano 2 is already attracting interest from companies working on different types of physical AI applications.

Cognex, Doosan Bobcat and Matic are among the companies adopting or exploring the new platform.

Matic Robotics plans to use Jetson Orin Nano 2 in home-cleaning robots, where the company says the technology can support conversational AI, gesture detection, precision mapping, semantic understanding and autonomous cleaning.

Matic cofounder and CEO Navneet Dalal said the company wants its robots to understand people and spaces while navigating changing home environments.

Jetson Technology Could Help Advance Drone Delivery

Drone delivery is another potential application.

Wing, the drone delivery company and Alphabet subsidiary, currently uses Jetson Orin Nano Super and NVIDIA’s software stack in its delivery drone fleet.

Wing plans to evaluate Jetson Orin Nano 2 for real-time AI perception and reasoning.

The company says greater responsiveness and energy efficiency could help make autonomous delivery drones faster and more dependable.

“Drone delivery depends on AI that can enable fast, reliable understanding of the real world,” Wing Head of Perception Dinuka Abeywardena said.

A Growing Hardware Ecosystem

NVIDIA is also working with a broad network of hardware and software partners around Jetson.

Companies including AAEON, ADLINK, Advantech, Aetina, Antmicro, Aptiv, Auvidea, AVerMedia, Connect Tech, Seeed Studio and others are developing carrier boards, hardware systems, customized AI software and reference solutions.

That ecosystem could help developers move from prototypes to commercial products more quickly.

When Will Jetson Orin Nano 2 Be Available?

NVIDIA says the Jetson Orin Nano 2 module and developer kit are expected to become available in the first half of 2027.

Until then, developers and robotics companies will likely continue testing the platform and evaluating how its improved performance and power efficiency can be used in real-world applications.

If NVIDIA’s performance claims translate into practical improvements, Jetson Orin Nano 2 could make sophisticated AI inference more accessible for smaller robots, drones and edge devices.

The bigger development is the continued shift of AI from cloud servers toward physical machines that can perceive, reason and act locally.