NVIDIA® Jetson AGX Xavier™ This has set a new benchmark for the computing density, energy efficiency, and AI reasoning ability of edge devices. It represents the next round of evolution for a new generation of intelligent machines with end-to-end autonomous capabilities.
Breakthroughs in the field of embedded applications
The size of the Jetson AGX Xavier is only 100 x 87 millimeters, and its performance is comparable to that of a large workstation, but its size is only 1/10 of that of a workstation,. Therefore, it is very suitable for autonomous machines such as delivery and logistics robots, factory systems, and large industrial UAVs.
Bringing performance to edge devices
The Jetson AGX Xavier is the world's first computer specifically designed for autonomous machines, capable of supporting visual ranging, sensor fusion, localization and map drawing, obstacle detection, and path planning algorithms crucial for the next generation of robots. The compact design brings GPU workstation level performance - up to 32 TOPS peak computing power and 750 Gbps high-speed I/O performance.
Flexible use and powerful functionality
The Jetson AGX Xavier product series provides high-performance self-service machines in efficient systems. The Jetson AGX Xavier comes with pre configured power modes of 10W, 15W, and 30W, respectively. The Jetson AGX Xavier Industrial Edition comes with pre configured power modes of 20W and 40W. These pre configured power modes can be switched during operation and can be customized according to specific needs.
JETSON AGX XAVIER suitable for different applications
The Jetson AGX Xavier series modules provide up to 32 TOPS of AI performance, and NVIDIA's rich set of AI tools and workflows can help developers quickly train and deploy neural networks.
Technical Specifications
| Jetson AGX Xavier series |
|
| AGX Xavier 64GB | AGX Xavier | AGX Xavier Industrial grade |
|
| AI performance | 32 TOPS | 30 TOPS |
|
| GPU | NVIDIA Volta architecture, equipped with 512 NVIDIA CUDA cores and 64 Tensor cores |
|
| CPU | 8 core NVIDIA Carmel Armv8.2 64 bit CPU
8MB L2 + 4MB L3 |
|
| DL accelerator | 2x NVDLA |
|
| Visual accelerator | 2x PVA |
|
| Security engine | - | 2x Arm Cortex-R5（synchronous） |
|
| RAM | 64GB 256 bit LPDDR4x
136.5GB/s | 32GB 256 bit LPDDR4x
136.5GB/s | 32GB 256 位 LPDDR4x（Support ECC）
136.5GB/s |
|
| Storage | 32GB eMMC 5.1 | 64GB eMMC 5.1 |
|
| Camera | Up to 6 cameras (Up to 36 can be supported through virtual channels) 16 channel MIPI CSI-2 | 8 channel SLVS-EC D-PHY 1.2 (up to 40 Gbps) C-PHY 1.1 (up to 62 Gbps) | Up to 6 cameras (Up to 36 can be supported through virtual channels) 16 channel MIPI CSI-2 D-PHY 1.2 (up to 40 Gbps) C-PHY 1.1 (up to 62 Gbps) |
|
| Video Coding | 4x 4K60 (H.265)
8x 4K30 (H.265)
16x 1080p60 (H.265)
32x 1080p30 (H.265)
30x 1080p30 (H.264) | 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 decoding | 2x 8K30 (H.265)
6x 4K60 (H.265)
12x 4K30 (H.265)
26x 1080p60 (H.265)
52x 1080p30 (H.265)
30x 1080p30 (H.264) | 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 4.0 | 8x SLVS-EC
3x USB 3.1
single channel UFS | 8x PCIe 4.0
3x USB 3.1
single channel UFS |
|
| Consumption | 10 watt | 15 watt | 30 watt | 20 watt | 40 watt |
|
| Network | 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 和 DSPK、GPIO |
|
| Specifications and dimensions | 100mm x 87mm 699 pin connector Integrated thermal conductivity plate |
|
