NVIDIA unveils Jetson Orin Nano 2 edge AI robotics computer

NVIDIA has announced the Jetson Orin Nano 2, a new entry-level edge AI robotics computer designed for drones, robots, and vision AI systems. The module features 78 TOPS of AI performance, an 8-core Arm CPU, and 8GB of LPDDR5X memory. While maintaining the same compact form factor as its predecessors, the new platform delivers twice the inference performance of the Jetson Orin Nano Super.

Efficiency is a primary focus of the new hardware. In a 15W operating mode, the Jetson Orin Nano 2 can achieve the same performance as previous models while consuming 40% less power. This improvement is intended to extend the battery life of mobile devices like delivery drones and autonomous robots. The platform is optimized to run memory-efficient large language models (LLMs) and vision language models (VLMs) locally, such as NVIDIA Cosmos, Nemotron, Gemma 4, and Qwen 3.

Industry partners are already engaging with the technology. AAEON has confirmed plans to integrate the module into its fanless embedded AI system roadmap via the BOXER-8XXXAI product line. Early adopters and evaluators include Doosan Bobcat, Cognex, Matic, and Wing Aviation, the latter of which plans to transition its delivery drone fleet to the new platform to enhance real-time AI perception. The module and developer kits are expected to be available in the first half of 2027.