Description
The AAEON BOXER-8742AI is a high-performance AI edge computing system powered by the NVIDIA® Jetson Thor platform, designed for demanding industrial and AI applications. It delivers advanced processing capabilities for AI inference, machine learning, and computer vision tasks in edge environments.
The system features multiple connectivity options including 5GbE LAN, four 1GbE LAN ports with PoE/PSE 802.3af support (up to 60W), and versatile serial and digital I/O interfaces such as RS-232/422/485, CAN FD, and DIO/GPIO. Expansion flexibility is provided through M.2 slots for 5G, Wi-Fi, and NVMe SSDs, enabling scalable storage and wireless connectivity. With a wide operating temperature range of -25°C to 60°C, the BOXER-8742AI is built for reliable, always-on operation in harsh industrial and edge computing environments.
Key Features
- NVIDIA® Jetson T4000™
- 5GbE LAN x 1
- 1GbE LAN x 4 with PoE/PSE 802.3af (Max. 60W)
- RS-232/422/485, CAN FD, DIO/GPIO
- M.2 expansion for 5G, Wi-Fi, and NVMe SSD
- -25°C – 60°C Operating Temperature
Specification
| Specification | Details |
| AI Accelerator | NVIDIA® Jetson T4000™ |
| CPU | 12-core Arm® Neoverse®-V3AE 64-bit |
| GPU | 1536-core NVIDIA Blackwell architecture GPU with 64 fifth-gen Tensor Cores |
| System Memory | 64 GB 256-bit LPDDR5X |
| Storage Device | M.2 2280 M-Key x1 for NVMe |
| Display Interface | HDMI 2.0 x1 |
| Ethernet | 5GbE LAN x1 (RJ-45), 1GbE LAN x4 with PoE 802.3af/at, Max 60W |
| I/O Ports | USB 3.2 Gen 2 Type-A x4 |
| DB-15 x1 for RS-232/422/485 & DIO x8 | |
| DB-9 x1 for RS-232 & CAN FD x2 | |
| Mini USB x1 for OS Flash | |
| Recovery Button x1, Reset Button x1, Power Button x1 | |
| External SMA Antenna x6, SIM Slot x2 | |
| Expansion | M.2 2280 M-Key x1 for NVMe |
| M.2 3042/3052 B-Key x1 for 4G/5G (USB 3.0 + PCIe) | |
| M.2 2230 E-Key x1 for WiFi/BT/NVMe | |
| Indicators | Power LED x1 |
| OS Support | NVIDIA Jetpack™ 7.0 and up |
| Power Supply | 2-pin 12V–24V DC Input Phoenix Connector |
| Mounting / Mechanical | TBD |
| Dimensions (W x D x H) | 8.26” x 6.46” x 2.91” (210mm x 164.2mm x 74mm) |
| Weight | Gross: TBD, Net: TBD |
| Operating Temperature | -25°C ~ 60°C (13°F ~ 140°F), with 0.7 m/s airflow, per IEC60068-2 |
| Storage Temperature | -40°C ~ 85°C (-40°F ~ 185°F) |
| Storage Humidity | 5% ~ 95% @ 40°C, non-condensing |
| Anti-Vibration | 3.5G, 5–500Hz (operating) |
| Anti-Shock | 50G peak acceleration, 11ms |
| Certification | CE / FCC Class A |
Side-by-side comparison for the AAEON BOXER-8742AI, AAEON BOXER-8741AI and BOXER-8740AI
| Specification | BOXER-8742AI | BOXER-8741AI | BOXER-8740AI |
| AI Accelerator | NVIDIA® Jetson T4000™ | NVIDIA® Jetson T5000™ | NVIDIA® Jetson T5000™ |
| CPU | 12-core Arm® Neoverse®-V3AE 64-bit | 14-core Arm® Neoverse®-V3AE 64-bit | 14-core Arm® Neoverse®-V3AE 64-bit |
| GPU | 1536-core NVIDIA Blackwell GPU, 64 fifth-gen Tensor Cores | Higher-performance GPU integrated in T5000™ | 2560-core NVIDIA Blackwell GPU with fifth-gen Tensor Cores |
| System Memory | 64GB 256-bit LPDDR5X | 128GB onboard LPDDR5X | 128GB 256-bit LPDDR5X |
| Ethernet / Networking | 5GbE LAN ×1, 1GbE LAN ×4 with PoE 802.3af (Max 60W) | QSFP28 ×1 for 4×25GbE, 5GbE LAN ×4, PoE 802.3af/at (Max 60W) | QSFP28 ×1 for 4×25GbE, 5GbE LAN ×1 |
| Expansion | M.2 for 5G, Wi-Fi, NVMe SSD | M.2 for 5G, Wi-Fi 6, NVMe SSD | M.2 for 5G, Wi-Fi / Wi-Fi 7, NVMe SSD |
| I/O | RS-232/422/485, CAN FD, DIO/GPIO | RS-232, RS-485 ×2, CAN FD ×2, DIO/GPIO, advanced MIPI CSI-2, I²C, SPI, UART, fan, audio, SMA antenna, SIM slots | USB 3.2 Gen 2 ×2, USB 2.0 ×2, Micro USB (OS flash), control buttons |
| Operating Temperature | -25°C ~ 60°C | -20°C ~ 60°C | 0°C ~ 40°C |
| Industrial Rating | Rugged edge AI | MIL-STD-810H vibration/shock, industrial-grade | Industrial-grade (TBD) |
| Use Case | Mid-range AI inference, edge AI tasks | High-performance AI workloads, multi-camera and high-bandwidth industrial applications | High-speed edge AI, networking-intensive AI inference, data-driven industrial applications |
