Connect Tech has unveiled its Gauntlet carrier board, purpose-built for the NVIDIA Jetson Thor T5000 module, marking a significant leap forward in deploying AI at the edge. Designed to turn Jetson Thor’s raw horsepower into practical, field-ready systems, Gauntlet marries rugged reliability with rich I/O and fast storage.

Jetson Thor, powered by NVIDIA’s Blackwell GPU architecture, delivers up to 2,070 FP4 TFLOPS of AI performance, paired with 128 GB LPDDR5X memory and a 14-core Arm Neoverse-V3AE CPU, all optimised for transformer, generative AI, and multi-sensor workloads. In essence, the Gauntlet + Jetson Thor combo delivers industrial-grade AI performance with unmatched sensor flexibility, an ideal foundation for deploying advanced physical AI systems in the real world.

NVIDIA Jetson carrier board

One of my fellow writers has already written about a Raspberry Pi carrier board, which can be a good reference for learning about this kind of product.

Gauntlet Carrier Board for NVIDIA Jetson Thor Specifications:

  • SoM support: NVIDIA Jetson Thor T5000
  • CPU/GPU: 14-core Arm Neoverse-V3AE CPU, NVIDIA Blackwell GPU
  • Memory: up to 128GB LPDDR5X (on Jetson Thor)
  • Storage:
    • 1x M.2 M-key NVMe (PCIe Gen3 x4)
    • 1x M.2 M-key NVMe (PCIe Gen3 x2)
    • MicroSD slot
  • Networking:
    • 2x 10GBASE-T Ethernet
    • 2x 1GbE
  • Camera interfaces (via 16-lane MIPI expansion):
    • GMSL3
    • GMSL2
    • FPD-Link III
    • SDI
    • MIPI CSI-2
  • Video output: DisplayPort 1.4a
  • Expansion:
    • 1x M.2 E-key (WiFi/BT)
    • 1x M.2 B-key (LTE/5G)
  • USB:
    • 2x USB 3.2 Gen1 Type-A
    • 1x USB 3.2 Gen1 Type-C (OTG supported)
  • Other I/O
    • 2x CAN bus
    • GPIO / PWM
    • I²C, SPI, UART
  • Power input: +30V to +60V DC
  • Operating temperature: -40°C to +85°C (industrial range)
  • Dimensions: 155 mm x 126 mm
Gauntlet Carrier Board without NVIDIA Jetson Thor T5000

Gauntlet ships as a production-ready, JetPack 7.0-pre-flashed system with a Linux OS, CUDA, cuDNN, TensorRT, and drivers for AI and vision workloads. It also works with NVIDIA Isaac and DeepStream SDKs for thermal assembly, enabling quick integration of Jetson Thor into machines and robots.

The Connect Tech Gauntlet carrier board for NVIDIA Jetson Thor does not have an official list price published on the company’s website, and interested customers are required to request a quote directly or through authorised distributors. Resellers such as WDL Systems currently list the board as out of stock and also indicate pricing is available upon request, while related accessories like the power supply are priced separately. Connect Tech announced that the Gauntlet would be ready for orders starting in June 2025, though actual availability may vary by distributor and supply channel.

By Editor

Founder of Embedded News Hub. I started this platform to share useful insights, make tech simple, or help people discover new ideas.

Leave a Reply

Your email address will not be published. Required fields are marked *

Verified by MonsterInsights