2.2 NVIDIA Jetson Module
What Is an NVIDIA Jetson Module
Jetson modules are NVIDIA System-on-Module (SoM) products designed for edge AI and robotics. Each module integrates the CPU, GPU, AI accelerators, memory, and high-speed interfaces into a compact hardware unit. A carrier board then exposes power, display, camera, network, and expansion interfaces for practical system integration.
Jetson Modules Used Across reComputer Jetson
| Typical Jetson Platform | Official Compute Reference | How to Think About It |
|---|---|---|
| Jetson Nano | 0.5 TFLOPS (FP16) | An early entry-level platform suitable for Linux, JetPack, and basic vision experiments |
| Jetson Xavier NX | 21 TOPS | Better suited to more complex vision and traditional edge AI workloads, and still relevant for many existing projects |
| Jetson Orin Nano | up to 67 TOPS | The current entry to mid-range mainstream option for learning, basic multi-camera vision, and lightweight generative AI |
| Jetson Orin NX | up to 157 TOPS | A compact high-performance option for robotics, multi-camera systems, and larger model inference |
| Jetson AGX Orin | up to 275 TOPS | Designed for higher-end autonomous systems and multi-sensor fusion workloads |
| Jetson AGX Thor | up to 2070 FP4 TFLOPS | A next-generation platform for physical AI and robotics, with metrics that extend beyond traditional INT8 TOPS |