
Nvidia has unveiled Jetson Orin Nano 2 for edge robotics.
It doubles compute while using 40% less power.
This enables real-time AI on devices like delivery drones.
What happened
Nvidia announced Jetson Orin Nano 2, a robotics computer for running AI at the edge. It doubles compute over the previous generation, with 78 trillion operations per second, 8GB of memory, and an eight-core CPU, while using 40% less power in 15-watt mode.
Why it matters
Smaller and more efficient frontier AI models can now run on compact hardware, bringing real-time intelligence to devices like drones. The on-device processing lets robots react instantly without waiting for cloud data, which is key for autonomous navigation and delivery.
What to watch
Partners including Cognex, Doosan Bobcat, Matic Robotics, and Wing Aviation are already working on integrating the chip. The module and developer kit will be generally available starting in the first half of 2027.
Ask the AI about this article →
The announcement of Jetson Orin Nano 2 comes as AI models become smaller and more efficient, allowing them to run on less powerful hardware. Nvidia is positioning this as a way to bring frontier-level AI to edge devices, enabling real-time responses that don't depend on a wireless connection to the cloud. For applications like drones, this on-device processing is critical to avoid latency in navigation and other time-sensitive tasks.
Wing's exploration of the chip for its delivery drones illustrates a practical use case. The company says the technology could lead to more responsive and energy-efficient drones that can navigate and avoid obstacles autonomously, ultimately making local deliveries faster and more reliable. This aligns with a broader trend of embedding AI directly into physical systems rather than relying on remote computing.
The jump in performance and efficiency over the previous generation, combined with smaller form factor, appears intended to make advanced AI practical for lightweight robotics. Whether this translates to wider commercial adoption may depend on how well these early integrations perform in real-world conditions.
AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.
Free · takes 30 seconds · unsubscribe anytime
Ask AI anything about this article. Q&As are published on this page for other readers too.
Banco BS2, a Brasil-based digital bank, is pursuing a foundation-first approach to scaling enterprise AI
McKinsey's 2026 State of AI survey of 1,719 professionals found that 37% of respondents attribute at least som…

Chinese automakers are moving aggressively into humanoid robotics, drawing on existing technology, supply chai…

Nvidia has guaranteed up to $105 billion of OpenAI's data center leases at the PORTS-Pike campus in Ohio

Google is launching Gemini Enterprise for Legal, a product that uses MCP connectors to link Google's AI model…

Cisco Systems Inc. and Nvidia Corp