NVIDIA Introduces Jetson Orin Nano Super: A $249 Compact Generative AI Computer

NVIDIA Introduces Jetson Orin Nano Super: A $249 Compact Generative AI Computer

NVIDIA has unveiled the Jetson Orin Nano Super, a budget-friendly AI developer kit that delivers 70 trillion operations per second, supporting generative AI models, robotics, and large language models (LLMs).

Key Points:

  • NVIDIA Jetson Orin Nano Super delivers 67 trillion operations per second
  • Costs just $249, a significant drop from its predecessor's $499 price
  • Compact and power-efficient (7W-25W), it supports advanced generative AI, LLMs, and robotics applications
  • The compact device can run large language models and advanced AI applications typically reserved for much larger systems

NVIDIA CEO Jensen Huang announced the Jetson Orin Nano Super from his home kitchen - a fitting sequel to his famous oven-based HGX unveiling during the pandemic. But while the setting was casual, the implications of this launch could be significant.

The Jetson Orin Nano Super delivers a 70% performance increase over its predecessor, with 67 INT8 TOPS of AI compute power and 50% more memory bandwidth at 102 GB/s. This tiny powerhouse represents NVIDIA's latest push to democratize AI development, particularly in robotics and edge computing.

As AI workloads increasingly rely on generative models, smaller, affordable devices are critical for democratizing AI development. The Jetson Orin Nano Super supports transformer-based models for computer vision, natural language processing, and multimodal AI, enabling advanced applications like robotics, AI agents, and vision-based systems on a budget.

Huang emphasized the device’s versatility, saying, “It runs everything that the HGX does … even large language models,” while remaining compact enough to fit in the palm of a hand. It can power everything from autonomous robots to edge AI workstations.

The Jetson Orin Nano Super comes equipped with NVIDIA's Ampere architecture GPU featuring 1024 CUDA cores and 32 tensor cores, alongside a 6-core Arm CPU. This hardware configuration enables it to handle multiple AI applications simultaneously, including transformer models and large language models that were previously the domain of much larger systems.

For existing Jetson Orin Nano users, NVIDIA is offering software updates that provide up to 1.7x improvement in generative AI inference performance. This upgrade path protects current investments while extending the platform's capabilities.

The device integrates with NVIDIA's broader AI ecosystem, including Isaac for robotics and Metropolis for vision AI, providing developers with a comprehensive toolkit for building and deploying AI applications. This ecosystem approach could prove crucial as the industry moves from task-specific models to more general-purpose foundation models.

The bottom line: At $249, the Jetson Orin Nano Super democratizes access to generative AI capabilities, offering hobbyists and professionals a powerful tool to build AI solutions in robotics, computer vision, and beyond. With enhanced performance and affordability, it’s a small but significant step in bringing generative AI to the edge.

Chris McKay is the founder and chief editor of Maginative. His thought leadership in AI literacy and strategic AI adoption has been recognized by top academic institutions, media, and global brands.

Let’s stay in touch. Get the latest AI news from Maginative in your inbox.

Subscribe