
NVIDIA Jetson Orin Nano 2: A new entry-level edge AI robot computer for developers and startups
Published by AINave Editorial • Reviewed by Ramit
NVIDIA has announced the Jetson Orin Nano 2, a robotics computer designed for entry-level edge AI that promises to double inference performance while cutting power consumption. For builders working on robots, drones, or vision systems, this means frontier-level AI capabilities in a compact, power-efficient module.
The Jetson Orin Nano 2 delivers 78 TOPS of AI compute, 8GB of memory, and an 8-core Arm CPU. According to NVIDIA, it achieves twice the inference performance of the Jetson Orin Nano Super through enhanced Tensor Cores and higher memory bandwidth, while maintaining the same form factor. In 15-watt mode, it uses 40% less power than the previous module for equivalent performance.
The module runs on NVIDIA's open software stack, which includes Jetson agent skills for rapid development. Supported open models include Nvidia Cosmos, Nvidia Nemotron, Gemma 4, and Qwen 3, optimized for memory-efficient edge inference.
Cognex, Doosan Bobcat, and Matic are among the first companies exploring the Nano 2. Matic Robots plans to use it in home cleaning robots for conversational AI and semantic understanding. Wing, Alphabet's drone delivery subsidiary, currently uses the Jetson Orin Nano Super and plans to evaluate the Nano 2 for real-time perception and reasoning. A broad ecosystem of partners including Advantech, ADLINK, Seeed Studio, and Aptiv are developing carrier boards and reference solutions to accelerate time-to-market.
The Jetson Orin Nano 2 module and developer kit are expected to be available in the first half of 2027, so builders cannot prototype with it today. Performance claims are based on NVIDIA's internal testing and should be verified independently. Pricing has not been announced. The module targets entry-level edge AI, so teams with higher compute needs may look at the recently announced Jetson T3000 (865 FP4 TFLOPS) or Jetson T2000.
For AI builders focused on edge deployment, the Jetson Orin Nano 2 represents a meaningful step in making frontier-class models practical on low-power devices. The combination of 78 TOPS, 8GB memory, and an open software stack should enable a new wave of robotics and edge AI applications once hardware ships in 2027.
FAQs
Sources
- Nvidia unveils Jetson Orin Nano 2 for entry-level edge AI
- NVIDIA Announces Jetson Orin Nano 2 Robotics... | NVIDIA Newsroom
- NVIDIA Unveils Jetson Orin Nano 2 to Redefine Entry-Level Edge AI
- Nvidia unveils Jetson Orin Nano 2 for entry-level edge AI
- NVIDIA Jetson Orin Nano 2 Boosts Robotics & Edge AI by 2x While...
- NVIDIA unveils Jetson Orin Nano 2 robotics computer - Robotics 24/7
- Nvidia unveils Jetson Orin Nano 2 for edge AI robots and drones
- Aptiv, Nvidia Collaborate on Jetson Orin Nano 2
- Huge News For Nvidia Stock Fans
- DFI Showcases NVIDIA Jetson Orin-Based Edge AI Platforms for Vision Applications at COMPUTEX 2026
- DFI Showcases NVIDIA Jetson Orin-Based Edge AI Platforms for Vision Applications at COMPUTEX 2026
- Edge AI systems support four GMSL2 cameras
- Nvidia unveils Jetson Orin Nano 2 for entry-level edge AI
- Nvidia unveils Jetson Orin Nano 2 robotics computer By Investing.com
- Nvidia's Next AI Bet Is Taking Shape
- Aptiv Accelerating Production-Ready Physical AI with Support for NVIDIA Jetson Orin Nano 2






















