Specifications
Module
Jetson Thor T5000
AI Performance
2,070 FP4 TFLOPS
GPU
2560-core Blackwell, 96 Tensor Cores (5th Gen)
CPU
14-core Arm Neoverse V3AE 64-bit
Memory
128GB 256-bit LPDDR5X
Networking
1× QSFP28 (4× 25GbE, 100Gbps), 4× 5GbE RJ-45
I/O
CAN FD × 2, RS-232/485, GPIO 8-bit, USB 3.2 Gen 2 × 4
Storage
M.2 2280 NVMe × 1
Expansion
M.2 B-Key 4G/5G, M.2 E-Key WiFi/BT
Operating Temp
-25°C ~ 60°C
Power
12-24V DC, 2-pin terminal block
Dimensions
240 × 164.8 × 80.1 mm
Certification
MIL-STD-810H, FCC/CE Class A
OS
JetPack 7.0+
Overview
The AAEON BOXER-8741AI is a fanless embedded AI box PC powered by the NVIDIA Jetson Thor T5000 module. With 2,070 FP4 TFLOPS from its 2560-core Blackwell GPU, 128GB LPDDR5X memory, and 100GbE aggregate networking via QSFP28, it delivers data-center-class AI inference at the rugged edge. Four independent 5GbE ports, dual CAN FD, and MIL-STD-810H ruggedization make it ideal for autonomous mobile robots (AMR), industrial inspection, and roadside AI deployment.
Key Benefits
Blackwell Architecture: 5th-generation Tensor Cores with FP4 support deliver breakthrough inference throughput for VLMs, LLMs, and multi-modal AI at the edge. 100GbE Fabric: QSFP28 port provides 4 independent 25GbE MACs for high-bandwidth sensor fusion and multi-camera vision pipelines. Industrial Rugged: -25°C to 60°C fanless operation with MIL-STD-810H shock/vibration certification, wall-mountable in compact 240×165×80mm footprint.
Request a Quote — AAEON BOXER-8741AI
QS Compute — NVIDIA Goldendisk Partner. Global B2B supply, 15-day sample lead time.
Get Your Quote →