NVIDIA announced the Jetson Orin Nano 2, a robotics computer designed to redefine entry-level edge AI by delivering frontier-class generative AI capabilities to millions of developers.

The new module delivers 2x the inference performance of its predecessor, the Jetson Orin Nano Super, while consuming 40% less power at the same performance level in 15-watt mode. This efficiency gain allows compact devices to run advanced models without increasing energy costs or form factor size.

Technical Specifications

The Jetson Orin Nano 2 features 78 trillion operations per second of AI compute, 8GB of memory, and an 8-core Arm CPU. These specifications enable the system to run large language models and vision language models optimized for memory-efficient edge inference.

Developers can utilize open models such as NVIDIA Cosmos, NVIDIA Nemotron, Gemma 4, and Qwen 3 to build applications for robots, delivery drones, and inspection systems. The module is expected to be available in the first half of 2027.

Ecosystem Adoption

More than 3 million developers are currently building on the NVIDIA robotics stack. Early adopters include Cognex, Doosan Bobcat, and Matic Robots. Wing, a drone delivery subsidiary of Alphabet, plans to evaluate the new hardware to advance real-time AI perception for its delivery fleet.

Matic Robots is adopting the Jetson Orin Nano 2 to enable conversational AI and autonomous cleaning capabilities in its home robots. The company aims to use the platform for real-time perception, interaction, and navigation in dynamic environments.

What the Numbers Show

The doubling of inference performance alongside a 40% reduction in power consumption indicates a significant improvement in performance-per-watt efficiency. This metric divergence suggests that NVIDIA is prioritizing thermal and energy constraints critical for battery-operated edge devices like drones and home robots, rather than solely maximizing raw compute throughput.

SpaceXAI will deploy NVIDIA Vera CPUs to accelerate its next generation of agentic AI applications. This collaboration brings the first CPU built specifically for AI agents to one of the world’s most ambitious AI deployments.

The move expands SpaceXAI’s infrastructure for Grok using the NVIDIA Vera Rubin platform. The architecture supports scaling toward gigawatts of computing capacity while extending optimized systems into space.

Agentic AI Infrastructure

Agentic AI applications increasingly rely on CPUs to orchestrate tools, execute code, process data and run simulations between model calls. SpaceXAI uses Vera to accelerate these workloads. This helps AI agents act faster while keeping GPUs fed and fully utilized.

NVIDIA Vera features 88 NVIDIA-designed Olympus cores, Spatial Multithreading technology and high-bandwidth LPDDR5X memory. The CPU delivers up to 1.2TB/s of bandwidth. It enables up to 1.8x faster task completion compared with x86 CPUs across agentic AI, reinforcement learning and data processing workloads.

"Vera gives us the CPU performance and memory bandwidth to run enormous amounts of orchestration, code and data processing while keeping GPUs doing what they do best," said Mike Nicolls, president of SpaceXAI.

Scaling With Vera Rubin

NVIDIA Vera Rubin is codesigned across compute, networking and software. The platform integrates NVIDIA accelerated computing, NVLink interconnect technology, Spectrum-X Ethernet networking, BlueField data processing and NVIDIA software.

This integrated architecture aims to deliver high performance and energy efficiency at scale. It drives down cost per token as SpaceXAI expands toward gigawatts of computing capacity. Vera Rubin provides a common architecture to efficiently scale next generation AI factories.

Orbital Computing Expansion

SpaceXAI plans to extend its use of NVIDIA accelerated computing into space. The first-generation Starmind AI satellite will be based on an optimized NVIDIA Vera Rubin NVL72 system.

Orbital computing imposes different constraints regarding power, thermal management, bandwidth, reliability and physical integration. NVIDIA and SpaceXAI are adapting the foundation to these requirements while preserving a common architecture and software ecosystem.

This delivers one computing foundation across environments. Vera CPUs accelerate AI agents on Earth. Vera Rubin powers the AI infrastructure behind Grok. NVIDIA accelerated computing extends into orbital AI infrastructure.