Overview
- NVIDIA announced the Jetson Orin Nano 2 on Tuesday, August 25, 2026, specifying 78 TOPS of AI compute, 8 GB of memory and an eight‑core Arm CPU and saying it delivers twice the inference performance of its predecessor while using about 40% less power at the same performance level.
- NVIDIA plans to ship a module and a developer kit in the first half of 2027 and will support the product with its open Jetson software stack and JetPack tooling so developers can run memory‑efficient large language and vision models on edge devices.
- Hardware and system partners including Cognex, Doosan Bobcat, Matic, Wing and a long list of carrier‑board vendors are building carrier boards, reference designs and production systems to help customers move prototypes into products.
- Aptiv said it will expand collaboration with NVIDIA to provide Yocto and enterprise Linux support for production deployments, a move that coincided with a notable intraday stock rise for Aptiv and a smaller uptick in NVIDIA shares.
- The change follows earlier Jetson upgrades that used software to raise performance, and the Orin Nano 2 combines silicon and microarchitecture improvements to broaden access to on‑device AI for the three million developers in the Jetson ecosystem and to shorten the path from lab prototypes to low‑power fielded robots and drones.