NVIDIA® Jetson AGX Xavier™ sets a new bar for compute density, energy efficiency, and AI inferencing capabilities on edge devices. It’s the next evolution in next-generation intelligent machines with end-to-end autonomous capabilities.
NVIDIA Jetson AGX Xavier Industrial delivers the highest performance for AI-embedded industrial and functional safety applications in a power-efficient, rugged system-on-module. It’s form-factor and pin-compatible with Jetson AGX Xavier, and with up to 20X the performance and 4X the memory of Jetson TX2i, customers can bring the latest AI models to their most demanding use cases. The Jetson AGX Xavier Industrial module is available for order now and will ship in late July 2021.
Features
Size
A BREAKTHROUGH IN EMBEDDED APPLICATIONS
At just 100 x 87 mm, Jetson AGX Xavier offers big workstation performance at 1/10 the size of a workstation. This makes it ideal for autonomous machines like delivery and logistics robots, factory systems, and large industrial UAVs.
Performance
TAKING PERFORMANCE TO THE EDGE
As the world’s first computer designed specifically for autonomous machines, Jetson AGX Xavier has the performance to handle visual odometry, sensor fusion, localization and mapping, obstacle detection, and path-planning algorithms that are critical to next-generation robots. Get GPU workstation-class performance with up to 32 TOPS of peak compute and 750 Gbps of high-speed I/O in a compact form factor.
Power
THE POWER OF FLEXIBILITY
The Jetson AGX Xavier series provides the highest level of performance for autonomous machines in a power-efficient system. Jetson AGX Xavier ships with configurable power profiles preset for 10W, 15W, and 30W, and Jetson AGX Xavier Industrial ships with profiles preset for 20W and 40W. These power profiles are switchable at runtime and can be customized to your specific application needs.
A JETSON AGX XAVIER FOR ANY APPLICATION
The Jetson AGX Xavier series of modules deliver up to 32 TOPS of AI performance, and NVIDIA’s rich set of AI tools and workflows lets developers train and deploy neural networks quickly.
Specifications
GPU | NVIDIA Volta architecture with 512 NVIDIA CUDA cores and 64 Tensor cores 20 TOPS (INT8) |
CPU | 8-core NVIDIA Carmel Arm® v8.2 64-bit CPU 8MB L2 + 4MB L3 |
DL Accelerator | 2x NVDLA 10 TOPS (INT8) |
Vision Accelerator | 2x 7-Way VLIW Vision Processor |
Safety Cluster Engine | Dual Arm Cortex-R5 in lockstep |
Memory | 32GB 256-bit LPDDR4x (ECC support) 136.5GB/s |
Storage | 64GB eMMC 5.1 |
CSI Camera | Up to 6 cameras (36 via virtual channels) 16 lanes MIPI CSI-2 D-PHY 1.2 (up to 40Gbps) C-PHY 1.1 (up to 62Gbps) |
Video Encode | 2x 4K60 (H.265/H.264) 6x 4K30 (H.265/H.264) 12x 1080p60 (H.265/H.264) 24x 1080p30 (H.265/H.264) |
Video Decode | 2x 8K30 (H.265) 4x 4K60 (H.265) 8x 4K30 (H.265) 18x 1080p60 (H.265) 36x 1080p30 (H.265) 24x 1080p30(H.264) |
UPHY | 8x PCIe Gen4 3x USB 3.1 Single Lane UFS |
Networking | 10/100/1000 BASE-T Ethernet |
Display | Three multi-mode DP 1.2a/e DP 1.4/HDMI 2.0 a/b |
Other I/O | USB 2.0 UART, SPI, CAN, I2C, I2S, DMIC & DSPK, GPIOs |
Power | 20W | 40W |
Mechanical | 100mm x 87mm 699 pin Molex Mirror Mezz Connector Integrated Thermal Transfer Plate |