NVIDIA Edge Platforms (Jetson, Orin, Thor)
NVIDIA Jetson devices are the world’s leading platform for GPU-accelerated AI computing at the edge. They deliver high-performance, low-energy computing for deep learning, computer vision and robotics, and the range of form factors and I/O interfaces makes Jetson the ideal choice for compute-intensive embedded projects. Every Jetson device runs the same NVIDIA CUDA software stack, so models trained in the data centre port directly to your edge device of choice. From research and prototyping through to full production deployment, there is a Jetson platform to support each step of taking AI to the edge.
The current line-up spans three tiers. The Jetson Orin Nano series (up to 67 TOPS) sets the baseline for entry-level edge AI in the smallest form factor, while the Orin NX series (up to 157 TOPS) and the AGX Orin series (up to 275 TOPS) power multiple concurrent AI inference pipelines for applications from manufacturing and logistics to retail and healthcare. At the top, the new Jetson AGX Thor series, built on the Blackwell GPU architecture, delivers up to 2,070 FP4 TFLOPS and 128 GB of memory for physical AI, humanoid robotics and generative AI at the edge, more than 7.5 times the compute of AGX Orin.
Each platform is available in two form factors. Development Kits include standard input/output ports like USB and HDMI along with on-board storage, making them ideal for evaluation and software development. Production modules strip away the I/O and are designed to slot into carrier boards, motherboards or custom builds for deployment at scale.
XENON can supply NVIDIA Jetson Development Kits and modules in bulk. For orders of 10 or more, or for special one-off projects, contact our team with your requirements and we will confirm lead times and provide a quote.