Jetson AGX Xavier is an embedded AI module for autonomous machines and edge inference systems. NVIDIA lists the 64GB family at up to 32 TOPS AI performance.1
This entry uses the published 32 TOPS module figure directly on the dense INT8 scale as 3.20e+13 operations per second in dataset notation.1
Total compute: 3.20e+13 operations per second.
AGX Xavier offers materially more on-device AI capability than entry-level Jetson modules while still fitting edge power and thermal envelopes. It remains below current Orin and Thor tiers.
NVIDIA. “Jetson AGX Xavier Series.” Accessed 2026. https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-agx-xavier/. ↩ ↩2