Specifications
Manufacturer
ASUS IoT
AI Module
NVIDIA Jetson Thor
T5000 or T4000
AI Performance
Up to 2070 TFLOPS (FP4)
Memory
128 GB LPDDR5X
CPU
14-core Arm (Thor module)
Video Output
HDMI 2.0
Networking
GbE + 10GbE RJ45
4× 25GbE links
Camera Support
Up to 16 GMSL cameras
Expansion
M.2 sockets (storage/wireless)
PCIe x4 / x8 slot
Power
12–60V DC wide input
Ruggedization
MIL-STD-810H grade
Form Factor
Compact, modular I/O
optional second stack
Target
Physical AI, robotics,
industrial AI, automation
Overview
The ASUS PE3000N is a rugged, deployment-ready edge AI supercomputer built on the NVIDIA Jetson Thor platform. Delivering up to 2,070 FP4 TFLOPS of AI performance with 128GB of built-in LPDDR5X memory, it brings advanced perception, reasoning and action to the edge for physical AI, robotics and industrial automation workloads.
Connectivity is engineered for multi-sensor systems: HDMI 2.0 video output, Gigabit and 10GbE RJ45 ports, four 25GbE links, M.2 sockets for storage and wireless expansion, and a PCIe x4/x8 slot. A modular I/O design supports up to 16 GMSL cameras, and an optional second stack doubles expansion capacity for the most demanding deployments.
MIL-STD-810H-grade construction with a 12–60V wide DC input makes the PE3000N suitable for in-vehicle and harsh-environment installations, positioning it as ASUS IoT's flagship for physical AI at the edge.
Request a Quote — ASUS PE3000N
QS Compute — global B2B supply of AI computing hardware. Price on request, stock availability confirmed on inquiry.
Get Your Quote →