NVIDIA Jetson Orin Nano 2 Delivers 78 TOPS for Enterprise Edge Robotics

The NVIDIA Jetson Orin Nano 2 is a new entry-level robotics computer for edge AI applications in robotics, drones and vision systems.

NVIDIA announced the Jetson Orin Nano 2 on August 25, 2026. The new robotics computer is set to redefine entry-level edge AI. It puts frontier-class generative AI performance in the hands of millions of developers. The platform targets robotics, drones, and vision AI applications. The timing coincides with the full production of the Groq 3 LPX. This creates opportunities for combined edge and cloud solutions. Developers can explore new use cases in physical AI.

Physical AI is moving from experimentation to deployment. Device manufacturers need partners who can help them scale successfully. In addition to powerful compute, reliable perception is important. Proven software platforms are necessary. The systems expertise required to operate safely and efficiently in dynamic environments is also key. The announcement addresses these requirements through hardware and ecosystem support.

What background led to the Jetson Orin Nano 2 launch?

The NVIDIA robotics stack has seen adoption by more than 3 million developers. These developers have built applications across various domains. The demand for real-time intelligence at the edge has grown. Small and medium frontier models have reached the accuracy of last year’s largest frontier models. This progress unlocks real-time intelligence for edge devices. The large developer base provides a foundation for rapid uptake of the new hardware.

The December 2025 licensing deal with Groq for $20 billion has enabled new inference technologies. The Groq 3 LPX is one result. It entered full production in August 2026. This timing aligns with the Jetson announcement. Enterprises can now access both edge and cloud inference capabilities. The deal strengthens NVIDIA's position in the inference market.

What technical features define the Jetson Orin Nano 2?

The Jetson Orin Nano 2 features 78 TOPS of AI compute. It includes 8GB of memory. An 8-core Arm CPU is integrated. The design maintains the same compact form factor as its predecessor. Improved Tensor Cores contribute to the performance gains. Higher memory bandwidth supports the enhanced capabilities. These elements combine to enable more advanced processing at the edge.

Inference performance is twice that of the Jetson Orin Nano Super. This holds in the same form factor. In 15-watt mode, the device consumes 40 percent less power to deliver the same performance as its predecessor. These specifications support energy-efficient operation in edge scenarios. Battery life extends in portable robotics and drone systems.

How does the Groq 3 LPX complement the edge platform?

The Groq 3 LPX inference accelerator is based on technology from the licensing deal. Groq is among the first to bring it to market. It boosts inference token generation for NVIDIA Vera Rubin NVL72. The chip will be deployed to a purpose-built AI inference cloud. This development supports high-performance inference needs. The production status indicates readiness for enterprise use.

Edge devices like the Jetson Orin Nano 2 can handle initial processing. More demanding tasks can route to cloud systems. The combination allows for optimized workloads. Enterprises benefit from this hybrid approach. Latency reduces for time-sensitive operations. Overall system throughput increases with the paired technologies.

Which partners support the Jetson Orin Nano 2?

Aptiv announced support for the Jetson Orin Nano 2. The company aims to help customers accelerate deployment of production-ready physical AI. Intelligent edge systems are a focus area. Aptiv provides the systems expertise for safe operation. The partnership covers perception and software integration needs.

Cognex is adopting the platform for vision AI. Doosan Bobcat is integrating it into equipment. Matic Robotics is using it for robotics platforms. Wing is deploying it in drone systems. These adoptions span multiple industries. The range of partners shows versatility across enterprise applications.

Key partners and their areas of focus with the Jetson Orin Nano 2.

Partner	Focus Area
Aptiv	Production-ready physical AI and intelligent edge systems
Cognex	Vision AI applications
Doosan Bobcat	Equipment integration
Matic Robotics	Robotics platforms
Wing	Drone operations

What are the implications for enterprise markets?

The new platform lowers the barrier for entry into advanced edge AI. Millions of developers can now access the performance needed for real-time reasoning. Energy efficiency improvements allow for broader deployment in power-constrained environments. The market for physical AI systems is positioned for expansion. Cost structures for edge deployments may improve over time.

Stakeholders in sectors such as manufacturing and logistics can implement more capable systems. Real-time decision making improves operational efficiency. The availability of supporting software and hardware from partners facilitates integration. This could lead to increased adoption rates over the coming years. Data processing at the edge enhances privacy controls.

The same form factor ensures compatibility with existing designs. Manufacturers do not need to redesign enclosures. This compatibility speeds up time to market for new products. The power savings further enhance the value proposition. Enterprise users gain flexibility in device selection.

What reactions have come from industry experts?

Deepu Talla, vice president of robotics and edge AI at NVIDIA, provided comments on the announcement. The statement addresses the capabilities of current models. It highlights the role of the new computer in enabling real-time intelligence. The remarks position the hardware as accessible to a wide developer audience.

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. 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.Deepu Talla, vice president of robotics and edge AI at NVIDIA

Jay Bellissimo, Senior Vice President and President, Intelligent Systems, Software and Services at Aptiv, also commented. The remarks emphasize the need for partners in scaling physical AI. Reliable perception and proven platforms are highlighted. The comments underscore the importance of systems expertise in dynamic settings.

The comments from industry leaders reflect confidence in the technology. The focus on energy efficiency aligns with sustainability goals. Real-time reasoning capabilities open new application areas. The developer community is expected to respond positively. These reactions provide validation for the product direction.

What developments are anticipated in the coming period?

The NVIDIA Jetson Orin Nano 2 module and developer kit are expected to be available in the first half of 2027. Developers can prepare applications in advance. Further integrations with partners are likely. The Groq 3 LPX will support additional inference workloads. The ecosystem continues to evolve with these releases.

The overall ecosystem for edge AI is expected to mature. More use cases in enterprise settings may emerge. The focus on physical AI deployment will continue. This announcement represents one step in that direction. Monitoring of adoption metrics will provide further insights.

1. Announcement of the Jetson Orin Nano 2 occurred on August 25, 2026.
2. Groq 3 LPX entered full production during the same month.
3. Partner support announcements followed the launch.
4. Availability of hardware is set for the first half of 2027.
5. Integration into enterprise systems will proceed thereafter.