NVIDIA 900-82888-0050-000 | Jetson AGX Xavier 64GB AI Compute Module, 32 TOPS, Volta GPU

Q: What is the difference between the 64GB and 32GB Jetson AGX Xavier modules?
A: The 64GB module doubles the memory capacity from 32GB to 64GB LPDDR4x, enabling storage and execution of multiple AI models simultaneously, more complex applications, and enhanced real-time processing pipelines. Both share the same GPU, CPU, and AI performance capabilities.

Q: Does this module require a carrier board?
A: Yes, the Jetson AGX Xavier module is a system-on-module that requires a compatible carrier board for operation. Options include NVIDIA developer kits, third-party carrier boards, or custom-designed carrier boards that provide power delivery, I/O connectivity, and thermal management.

Q: What AI frameworks and software are supported?
A: The module is supported by NVIDIA JetPack SDK, which includes Linux BSP, CUDA, cuDNN, TensorRT, and support for popular AI frameworks including TensorFlow, PyTorch, and ONNX. It also supports NVIDIA DeepStream SDK for video analytics and Isaac SDK for robotics applications.

Q: What are typical use cases for the 64GB variant?
A: The 64GB module is ideal for autonomous machines, industrial robotics, optical inspection systems, edge AI servers running multiple models, smart city infrastructure, medical imaging devices, and any application requiring substantial on-device memory for complex AI workloads.

Q: What interfaces and connectivity does the module support?
A: The module provides PCIe Gen4, USB 3.1, Gigabit Ethernet, MIPI CSI-2 camera interfaces, I2C, SPI, UART, GPIO, CAN, I2S audio, and DisplayPort/HDMI video output through the carrier board connector, offering 750 Gbps of total I/O bandwidth.