Nvidia on Tuesday unveiled the Jetson Orin Nano 2, a compact robotics computer designed for entry-level edge AI applications. The Santa Clara-based chipmaker positioned the device as a successor to the Jetson Orin Nano Super, targeting developers building small-form-factor systems for robotics and autonomous machines.
The module offers 78 trillion operations per second of AI compute alongside 8GB of memory and an 8-core Arm CPU. Nvidia said the Jetson Orin Nano 2 delivers twice the inference performance of its predecessor while consuming 40% less power when operating in a 15-watt mode. The company maintained the same compact form factor as earlier models to support existing hardware designs.
Nvidia’s robotics stack has already attracted more than 3 million developers, and the Jetson Orin Nano 2 supports large language models and vision language models optimized for edge inference, including Nvidia Cosmos, Nvidia Nemotron, Gemma 4 and Qwen 3. The company highlighted early interest from companies evaluating the system for commercial applications.
Wing, an Alphabet subsidiary developing drone delivery systems, said it plans to evaluate the Jetson Orin Nano 2 for its fleet, which currently uses the Jetson Orin Nano Super. Dinuka Abeywardena, head of perception at Wing, said the new module could enable faster, more energy-efficient drones by improving real-world scene understanding.
Matic Robotics, which builds home cleaning robots, will adopt the system to support conversational AI, gesture detection and autonomous cleaning capabilities. Navneet Dalal, cofounder and CEO of Matic Robotics, said the Jetson Orin Nano 2 allows the company to run state-of-the-art AI models at the edge within a compact platform.
Cognex and Doosan Bobcat are among other companies exploring the new system, according to Nvidia. The Jetson Orin Nano 2 and its developer kit are expected to ship in the first half of 2027.
