With Jetson Orin Nano 2, NVIDIA doubles inference performance while cutting power consumption. Embedded computing specialist Syslogic is announcing several rugged computers based on the new edge AI module.
For years, NVIDIA Jetson Nano has provided a particularly compact and energy-efficient entry point into the Jetson platform. Now, NVIDIA is launching the next generation with the NVIDIA Jetson Orin Nano 2. The new module delivers up to 78 TOPS (trillion operations per second) of AI performance and features 8 GB of memory and an 8-core Arm CPU.
Compared with the Jetson Orin Nano Super, NVIDIA has doubled inference performance. This is made possible, among other improvements, by enhanced Tensor Cores and increased memory bandwidth. As a result, current large language models (LLMs) and vision language models (VLMs) can also be run directly at the edge.
In addition to increased AI performance, the improved energy efficiency is particularly relevant for industrial applications. In 15-watt mode, the Jetson Orin Nano 2 consumes 40 percent less power than its predecessor at the same performance level.
Lower power consumption reduces thermal stress on the electronics and facilitates the development of compact, passively cooled embedded systems. At the same time, it provides additional thermal headroom, which is particularly beneficial for applications operating across extended temperature ranges. This makes the Jetson Orin Nano 2 well suited for mobile machinery, autonomous vehicles, and robotics applications where compact dimensions, high reliability, and low power consumption are critical.
Syslogic is announcing several IP67/IP69-rated versions of its Rugged Computer RPC RS based on the new NVIDIA Jetson Orin Nano 2. Michael Jung, Product Manager at Syslogic, says: “In the first half of 2027, we will launch the particularly compact AI Rugged Computer RPC RS with Jetson Orin Nano 2.”
In terms of performance, the new systems will serve as the entry point into Syslogic’s NVIDIA Jetson portfolio. In terms of ruggedness and longevity, however, they will meet the same demanding standards as the company’s more powerful rugged computers. The new devices will be designed for an extended temperature range of –25°C to +70°C and will feature Syslogic’s proven rugged housings.
The new embedded computers are particularly well suited for computer vision and inference tasks such as person and object detection. Typical applications include mobile machinery, autonomous vehicles, robotics, and intelligent vision systems.
The NVIDIA Jetson Orin Nano 2 module is expected to become available in the first half of 2027. Syslogic also plans to launch its Jetson Orin Nano 2-based rugged computers in the first half of 2027.
The NVIDIA Jetson Orin Nano 2 is a compact edge AI platform for robotics, computer vision, autonomous systems, and other AI-enabled embedded applications. It combines an NVIDIA Ampere GPU with 1,536 CUDA cores, an 8-core Arm Cortex-A78 CPU, and 8 GB of LPDDR5X memory. With up to 78 TOPS of AI performance, the Jetson Orin Nano 2 brings powerful AI computing to the edge in a compact form factor.
The Jetson Orin Nano 2 delivers up to 78 TOPS of INT8 AI performance. Compared with the previous Jetson Orin Nano, it features more CUDA cores, additional CPU cores, and higher memory bandwidth of up to 120 GB/s. Depending on the workload, NVIDIA states that the Jetson Orin Nano 2 can deliver up to twice the inference performance of its predecessor.
The Jetson Orin Nano 2 brings a significant performance and efficiency boost while retaining the compact form factor of the Jetson Orin Nano family. Compared with the previous 8 GB model, the Jetson Orin Nano 2 increases the number of CUDA cores from 1,024 to 1,536 and the CPU cores from six to eight. It also increases memory bandwidth from 102.4 GB/s to 120 GB/s. At comparable performance, it can reduce power consumption by up to 40 percent.
The Jetson Orin Nano 2 is designed for AI inference at the edge, making it well suited for computer vision, object and person detection, robotics, autonomous machines, and intelligent embedded systems. Typical applications include mobile machinery, autonomous vehicles, AMRs and other robotic systems where AI models need to process sensor data locally and in real time.
The Jetson Orin Nano 2 combines high AI performance, low power consumption, and a compact form factor with the NVIDIA Jetson software ecosystem. This enables developers to run AI models directly on the edge without relying on cloud infrastructure. For industrial applications, the improved performance per watt also makes it easier to integrate powerful AI computing into compact, fanless, and rugged embedded systems.
