NVIDIA is pushing the boundaries of artificial intelligence by bringing sophisticated AI capabilities to the edge with its latest innovation, the Jetson Orin Nano 2. This new robotics computer is designed to empower drones, robots, and vision systems with on-device generative AI, a significant leap from traditional cloud-dependent processing.
The core of NVIDIA’s announcement hinges on the remarkable advancements in the performance of small and medium-sized AI models. Previously, such accuracy was only achievable with the most powerful, large-scale “frontier” models. Now, these more compact models are delivering comparable results, making advanced AI accessible on edge devices.
“The Jetson Orin Nano 2 computer puts that breakthrough within reach of millions of developers, delivering the performance and energy efficiency needed for real-time reasoning at the edge,” stated Deepu Talla, VP of Robotics and Edge AI at NVIDIA. This capability allows edge hardware to interpret complex data like language and images in real-time, enabling autonomous systems to act decisively without latency. This is crucial for applications such as delivery drones, industrial inspection drones, and sophisticated vision AI systems that demand both high performance and low power consumption.
**NVIDIA Jetson Orin Nano 2: Performance and Efficiency Unpacked**
The Jetson Orin Nano 2 boasts an impressive 78 trillion operations per second (TOPS) of AI compute power, coupled with 8GB of memory and an eight-core Arm CPU. NVIDIA has engineered this board to significantly enhance AI and video processing capabilities while maintaining a low cost and minimal power footprint.
Compared to its predecessor, the Jetson Orin Nano Super, the new board offers double the inference performance. This substantial improvement is attributed to enhancements in its Tensor Cores and increased memory bandwidth, all within the same compact form factor. Furthermore, the Jetson Orin Nano 2 demonstrates remarkable power efficiency, consuming 40% less power than the Orin Nano Super when operating in its 15-watt mode, while matching its performance levels.
Underpinning this hardware is NVIDIA’s robust open-source software stack, which includes Jetson agent skills and the broader Jetson AI ecosystem. The Jetson Orin Nano 2 is specifically optimized for running large language models (LLMs) and vision-language models (VLMs) that are memory-efficient for edge deployment. NVIDIA highlights its own Cosmos and Nemotron models, as well as popular open-source options like Gemma 4 and Qwen 3, as examples of models readily deployable on this new hardware.
**Pioneering Edge AI Applications with Industry Partners**
NVIDIA has already engaged early adopters, including Cognex, Doosan Bobcat, and Matic, to explore the potential of the Jetson Orin Nano 2 in real-world physical AI applications. With over three million developers already building on NVIDIA’s robotics platform, the company anticipates a surge in edge AI integration across various sectors, from home robots and advanced vision systems to drones and specialized hardware.
Wing, Alphabet’s drone delivery subsidiary, currently utilizes Jetson Orin Nano Super and NVIDIA’s software for its extensive drone fleet. The company is actively evaluating the Jetson Orin Nano 2 to further enhance its real-time AI perception and reasoning capabilities, aiming to expedite and enhance the safety of deliveries from local businesses to residential areas.
Dinuka Abeywardena, Head of Perception at Wing, emphasized the critical role of AI in drone delivery: “Drone delivery depends on AI that can enable fast, reliable understanding of the real world. Wing is exploring Jetson Orin Nano 2 to give us a path to more responsive, energy-efficient drones that can help make deliveries quicker and more dependable for customers.” While Wing is currently in an evaluation phase with the Jetson Orin Nano 2, a public timeline for production deployment on their delivery flights has not yet been announced.
Matic Robots, a company focused on consumer robotics, is integrating the Jetson Orin Nano 2 into its home cleaning robots. This integration will enable Matic to incorporate advanced features such as conversational AI, gesture detection, precise environmental mapping, and semantic understanding of home spaces, all while facilitating autonomous cleaning routines.
Navneet Dalal, Co-founder and Chief Executive of Matic Robots, elaborated on the impact: “Home robots need to understand people, map spaces precisely, understand the layout of objects and spaces, and clean autonomously in dynamic and constantly changing environments. With Jetson Orin Nano 2, Matic can run state-of-the-art AI models at the edge in a compact home robotics platform built for real-time perception, interaction, and navigation.”
**A Thriving Ecosystem: Carrier Boards and Reference Designs**
Complementing the Jetson Orin Nano 2, NVIDIA has announced a wide array of hardware partners developing carrier boards and comprehensive hardware systems. Manufacturers such as AAEON, ADLINK, Advantech, and Aetina are among those creating specialized hardware solutions.
Further extending the ecosystem, companies like Antmicro, Aptiv, Auvidea, and AVerMedia are focusing on customized AI software and reference designs. This collaborative effort, alongside contributions from Chuanglebo, Connect Tech, ForeCR, and JWIPC, aims to accelerate time-to-market for developers.
The list of partners dedicated to supporting customers includes Neurealm, Plink, Realtimes, RidgeRun, RS, Seeed Studio, Tauro Tech, Twowin, TZTEK, and YUAN.
“Frontier intelligence has reached the edge. Frontier models that used to run inside data centers last year can now run in real time on entry-level Jetson systems,” Talla concluded. “With Jetson Orin Nano 2, robots and edge systems can run leading language and vision models locally and in real-time, opening up applications that weren’t possible before.” This democratization of advanced AI through edge computing signifies a new era for autonomous systems and intelligent devices.
Original article, Author: Samuel Thompson. If you wish to reprint this article, please indicate the source:http://aicnbc.com/25151.html
