Technical specifications for the NVIDIA Jetson Xavier NX System-on-Module, detailing Volta GPU, Carmel CPU, memory, storage, and interface capabilities.
The NVIDIA Jetson Xavier NX is a System-on-Module (SoM) designed to bring AI supercomputer performance to the edge in a compact form factor smaller than a credit card. It integrates the NVIDIA Volta GPU architecture with 64-bit operating capability, advanced multi-function video and image processing, and NVIDIA Deep Learning Accelerators. The module supports compute performance up to 14 TOPS at 10W or 21 TOPS at 15W, enabling the parallel execution of multiple neural networks and simultaneous processing of data from high-resolution sensors.
| AI Performance | Up to 21 TOPS (INT8) at 15W |
|---|---|
| GPU Cores | 384 CUDA cores, 48 Tensor cores |
| CPU Architecture | ARMv8.2 (64-bit) heterogeneous multi-processing |
| CPU Cores | 3x dual-core CPU clusters (six NVIDIA Carmel processor cores) |
| Memory | 8 GB 128-bit LPDDR4x DRAM |
| Storage | 16 GB eMMC 5.1 Flash Storage |
| Networking | 10/100/1000 Gigabit Ethernet |
| Display Interfaces | eDP 1.4a, DP 1.4, HDMI 2.0a/b |
| Module Size | 69.6 mm x 45 mm |
| Operating Temperature | -25°C – 90°C |
The maximum AI performance is up to 21 TOPS (INT8) when operating at 15W.
The Volta GPU includes 384 CUDA cores and 48 Tensor cores.
At 15W, the CPU cores can operate at a maximum frequency of 1900 MHz.
The module size is 69.6 mm x 45 mm with a 260 pin SO-DIMM connector.