MIC-742​-AT

AI Inference System Accelerated by NVIDIA Jetson Thor Module

Manufacturer: Advantech
Manufacturer: Advantech
  • Industrial platform Edge AI based on NVIDIA Jetson Thor

  • Up to 2070 TFLOPS (FP4) - performance for AI and robotics

  • Jetson T5000 (Blackwell + Tensor Cores 5th gen.)

  • Service 8 GMSL2 (FAKRA) cameras - advanced vision

  • QSFP28 (4×25GbE) - high-performance communications

  • CANbus, I2C, LTE, WiFi - industrial systems integration

  • Platform for robotics, autonomy and video analytics

.

Product out of stock

Free shipping from €300

Promocja cenowa na model HDR-15-5

Product intended for professional use only
MIC-742​-AT

MIC-742​-AT

Description

 
 
Edge AI Computing / Robotics / Vision

MIC-742-AT - an AI platform designed for systems that need to see and respond

Advantech MIC-742-AT is an edge AI computer built around the NVIDIA Jetson Thor, designed for robotics, autonomy and multi-channel perception systems. It combines very high computing power with the interfaces needed to integrate cameras, sensors and communication infrastructure.

AI Performance
2070 TFLOPS
FP4
Vision
8x GMSL2
mini FAKRA
Networking
QSFP28
4 x 25GbE
Industrial I/O
CAN / I2C
robotics ready
NVIDIA Jetson Thor

New generation edge AI for the physical world

Jetson Thor is not just a more powerful embedded platform. It is an architecture prepared for systems that need to simultaneously analyse images, combine data from multiple sensors and make decisions locally.

With performance of up to 2070 TFLOPS (FP4) the platform can handle complex AI models, perception pipelines and control algorithms without moving computation to the cloud.

why it matters
  • less latency in autonomous systems
  • greater independence from cloud infrastructure
  • local analysis of image and sensory data
  • greater control over data security

Designed for perception and sensor fusion

MIC-742-AT is built around applications where AI needs to work on multiple inputs simultaneously: cameras, control interfaces and industrial or mobile sensor data.

01

Multicamera vision

Supports 8 channels of GMSL2.0 from the mini FAKRA allows the construction of advanced 360° perception, inspection and real-time video analysis systems.

  02

Fast communication

The QSFP28 (4 x 25GbE) interface enables the transfer of large volumes of data, integration with network infrastructure and the construction of distributed edge AI systems.

03

Device integration

CANbus, I2C, 5GbE, USB 3.2, WiFi and LTE facilitate integration with robots, sensors, controllers and mobile devices.

Applications

Edge AI where a decision must be made immediately

mobile and humanoid robots
autonomous vehicles and ADAS
intelligent image analysis
edge AI in industry and infrastructure
NEW GENERATION PLATFORM

MIC-742-AT as an AI platform for robotics, autonomy and advanced vision

If you are looking for an edge AI computer based on NVIDIA Jetson Thor that combines very high computational performance with interfaces for cameras, sensors and control systems, the Advantech MIC-742-AT is the platform designed for exactly such implementations. It is the solution for modern applications in the areas of robotics, autonomous systems, sensor fusion, image analysis and edge computing.

Technical Specification

Process
Model NVIDIA Jetson T5000
CPU 14 core Arm Neoverse V3AE (64-bit) SMP CPU architecture
L1 / L2 / L3 Cache L1 (I, D) per core: 64KB + 64KB, L2 1MB, L3 16MB
GPU 2560 NVIDIA CUDA cores, 96 5th Gen Tensor cores, MAXN: 1.57 GHz
Memory 128GB LPDDR5X
Ethernet
RJ45 1 x 10/100/1000/2500/5000 Mbps Ethernet
QSFP28 4 x 25GbE
I/O
Display HDMI (Max. resolution 3840x2160 @ 60Hz)
USB 4 x USB 3.2 Gen 2
OTG USB 1 x Micro USB
Console 1 x microUSB console for debug (UART to USB)
Audio 1 x audio speaker out, MIC in on board
CANbus 4 x CAN FD
GMSL (optional) 4 x MIPI 4 lanes (Support 8-ch GMSL2.0 with mini FAKRA connectors)
I2C 1 x I2C
SIM card 1 x Nano SIM on side of board
Expansion
M.2 E-key 1 x 2230 E-key (Signal: PCIe x1 + USB2.0)
M.2 B-key 1 x 3052/3042 B-key (Signal: USB3.0 + USB2.0)
SATA 2 x SATA connector
TPM 1 x TPM (onboard)
Storage
M.2 M-key 1 x 2280 M-key NVMe (PCIe x4)
Pre-installed SSD 1TB SSD
Power
Power Supply Power input: 19–36V
Power Type Terminal Block 2 Pin
Internal pin header 3 x SMART FAN
Environment
Operational Temperature -10 ~ 60 °C with 0.7 m/s air flow
Operating Humidity 95% @ 40 °C (non-condensing)
Vibration 3 Grms @ 5 ~ 500 Hz, random, 1 hr/axis
Mechanical
Dimensions (W x D x H) 217 x 76.4 x 184.5 mm
Weight 3kg
Installation Wall mount
Software & Certifications
Operating System NVIDIA Jetpack 7.0
Certifications CE/FCC/CCC/KC/KCC/BSMI/UL (No RED certification)

 

Recommended Products

Contact an Elmark specialist

Have questions? Need advice? Call or write to us!