The Jetson AGX Xavier 8GB Module is an accelerator for use in AI and deep-learning applications and is capable of up to 20 TOPS. The module is centered on the 384-Core NVIDIA Volta GPU, which includes Tensor Cores for accelerated computation of AIs designed around TensorFlow. The module also integrates a six-core Arm V8.2 64-bit Carmel CPU, 8-GB 256-bit LPDDR4x, 32-GB eMMC 5.1 flash storage, and a seven-way VLIW vision-accelerator processor. The module has voltage inputs of 5 V and 9–20 V and draws a power between 10 and 20 W. The module includes a range of connectivity options via external expansion, including 16× CSI-2 lanes, Gigabit Ethernet, HDMI, DP 1.2, 8× PCIe Gen 3, USB 3.1, UART, GPIO, I2C, I2S, and SPI. The module is 87 × 100 mm, has a 699-pin Molex Mirror Mezz Connector, and can be further prototyped with the Jetson AGX Xavier Developer Kit. The Jetson AGX Xavier module supports NVIDIA JetPack SDK, Linux OS, NVIDIA CUDA, cuDNN, and TensorRT.  

Drawings/Package

   

Key Features

• 384-Core NVIDIA Volta GPU

• Tensor Cores

• Six-core Arm V8.2 64-bit processor

• Minimum power draw of 10 W

• 8 GB of 256-bit LPDDR4x

General Applications

• AI

• Deep learning

• Object recognition

• Automation

• Speech recognition

Evaluation Board

YES — The Jetson AGX Xavier can be evaluated with the 945-82972-0000-000 available from Arrow.