NVIDIA Jetson AGX Thor is the next-generation compute platform for physical AI, designed for humanoid robots, industrial manipulators and autonomous vehicles. The module delivers 2070 TFLOPS at FP4 precision via a Blackwell GPU with second-generation Transformer Engine, a 14-core Arm Neoverse-V3AE CPU and 128 GB of LPDDR5X memory at 273 GB/s.
Thor delivers roughly a 7.5x performance leap over Jetson AGX Orin and 3.5x better energy efficiency, enabling on-board execution of large VLA (Vision-Language-Action) and generative world models. Power draw is configurable in the 40-130 W range, and the interface set includes QSFP 4x25GbE, USB 3.2, PCIe Gen5, CAN-FD and HDMI 2.0b.
AI System-on-Chip (SoC) · serves as: High-level compute, AI Inference, Computer vision processing, Embedded main processor.
Which group NVIDIA Jetson AGX Thor belongs to and how it is built
Compute Modules is a subcategory of hardware components that provide processing power for robotic systems. It encompasses onboard computers, single-board computers (SBCs), AI accelerators, embedded processors, GPU/NPU compute modules, and other units responsible for processing sensor data and executing control logic. These modules form the foundation of modern autonomous, humanoid, and perception-capable robots.
Component type encompassing System-on-Chip designs used as primary compute units in embedded devices, edge AI, and robotics applications.
Basic physical properties of NVIDIA Jetson AGX Thor — dimensions, weight and materials
