NVIDIA Jetson Orin Nano 2: 78 TOPS and 2x the Inference of Its Predecessor, Arriving First Half of 2027

NVIDIA announced the Jetson Orin Nano 2, a system-on-module and developer kit delivering 78 TOPS of AI compute—double the inference throughput of its predecessor—and consuming 40 percent less power at 15 watts, with availability expected in the first half of 2027. The module targets edge robotics, drones, and industrial devices, with early adopters including Cognex, Doosan Bobcat, Matic, and Wing.

NVIDIA has announced the Jetson Orin Nano 2 https://nvidianews.nvidia.com/news/nvidia-announces-jetson-orin-nano-2-robotics-computer-to-redefine-entry-level-edge-ai , an updated system-on-module and developer kit engineered to run compact foundation models and real-time computer vision workloads at the edge. The release targets embedded robotics, unmanned aerial vehicles, and industrial edge devices that require local inference capabilities without relying on continuous cloud connectivity. “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,” said Deepu Talla, vice president of robotics and edge AI at NVIDIA. “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.” Efficiency Gains The Jetson Orin Nano 2 delivers a substantial performance uplift over the original Orin Nano, providing 78 TOPS trillion operations per second of AI compute, effectively doubling the processing throughput of its predecessor. The module integrates an eight-core CPU alongside 8 GB of unified memory, packaged within a compact form factor suitable for space-constrained deployments. Power efficiency has also improved. In its 15-watt power mode, NVIDIA says the module consumes 40 percent less power than the prior generation while delivering the same performance. This thermal and power profile is designed for lightweight robotics, aerial drones, and mobile industrial platforms where battery capacity and payload constraints limit system design. By executing real-time vision processing and inference directly on the device, the hardware eliminates reliance on high-bandwidth wireless uplinks for core perception loops, ensuring deterministic response times and continuous operation in disconnected or signal-degraded environments. Edge Deployment Cognex, Doosan Bobcat, and Matic are among the first to adopt and explore the Jetson Orin Nano 2, with Wing lined up on the drone side. Jetson is also threading into larger deployments, from Amazon Robotics’ warehouse automation pipeline https://www.storagereview.com/news/aws-and-nvidia-to-deploy-2-million-more-gpus-in-2027-2028-bringing-vera-cpus-and-custom-nvhbm-to-trainium to NVIDIA’s broader physical AI push https://www.storagereview.com/news/nvidia-and-japan-launch-27500-gpu-vera-rubin-ai-factory-as-physical-ai-push-spans-every-industry . In aerial logistics, Wing, the Alphabet drone delivery subsidiary, already flies Jetson Orin Nano Super in its delivery fleet and plans to evaluate the Orin Nano 2 for onboard perception. “Drone delivery depends on AI that can enable fast, reliable understanding of the real world,” said Dinuka Abeywardena, head of perception at Wing. The increased compute density and power efficiency of the Orin Nano 2 are intended to support real-time sensor processing, obstacle avoidance, and path routing directly on the aircraft. Availability NVIDIA expects the Jetson Orin Nano 2 to be available as both a standalone production module and a developer kit in the first half of 2027.