Lantronix SoM brings AI and ML to edge

  • April 2, 2025
  • Steve Rogerson

California-based Lantronix has announced a system-on-module (SoM) to provides low-power, on-device AI and machine-learning (ML) capabilities, simplifying designs and helping developers bring edge products to market.

The production-ready Open-Q 8550CS module is powered by the Qualcomm Dragonwing QCS8550 processor.

It is designed to meet the higher AI and ML requirements of extreme edge computing, including video and AI applications such as video collaboration, video transcoding, cameras and integration with edge AI gateways. As with all Lantronix’s embedded compute technology, this platform comprises hardware, software, device management and services, helping users get to market faster.

It is suitable for developing industrial edge AI products, including drones, controllers, robotics and industrial handheld devices for industries, including smart warehousing, manufacturing, transportation, logistics and retail.

“Qualcomm’s 15-year strategic collaboration with Lantronix supports our mutual goal of delivering integrated, collaborative options to elevate the success of IoT, edge AI and ML technologies to drive the development of advanced edge applications,” said Suri Maddhula, vice president at Qualcomm.

Mathi Gurusamy, chief strategy officer at Lantronix, added: “With the support of Qualcomm, Lantronix is driving seamless AI innovation at the edge, empowering developers to harness embedded computing and IoT for cutting-edge, industrial-grade options. Together, we’re transforming the impossible into reality.”

The SoM features an on-device AI engine supporting the higher AI and ML requirements for extreme edge computing, including edge devices, servers and AI boxes. Security features include trusted management engine, hypervisor, secure processing unit and DDR encryption.

Features include:

  • Low power consumption with a 4nm process
  • Kryo Octa-core CPU up to 3.2GHz and Adreno A740 GPU
  • Dual eNPU delivering 48 INT8, 12 FP16 Tops
  • Enterprise-level connectivity with Wifi 7 MU-MIMO supporting up to 5.8Gbit/s
  • Up to 8Gbyte LPDDR5 RAM plus 128Gbyte UFS flash
  • Android 13 and Linux Yocto Kirkstone
  • Dedicated computer vision engine
  • Multiple Mipi camera and display ports
  • Multiple high-speed connectivity options
  • Support for Qualcomm Sensing Hub 3.0

Benefits include the ability to enhance video conferencing meeting experiences, automated guided vehicle pathing, smart camera image quality and edge AI box scalability with its octal-core computing capabilities and 48 AI Tops tensor performance.

The SoM (www.lantronix.com/products/open-q-8550cs-som) can perform complex 3D rendering and computer vision tasks with an Adreno 740 GPU supporting ray tracing, Open GL ES, Vulkan and Open CL profiles and 4K240 and 8K60 video decoding and 4K120 and 8K30 encoding. Edge AI boxes can be connected leveraging 2.5G and 10G Ethernet ports.

The firm’s Open-Q 8550CS SoM development kit (www.lantronix.com/products/open-q-8550cs-som-development-kit) facilitates evaluation of the SoM’s key features, such as the low-power AI subsystem with a dedicated DSP and AI accelerator supporting always-on audio, sensors, contextual data streams and an always-on camera.

The kit supports the evaluation of C-Phy and D-Phy Mipi CSI and GMSL cameras, dual Mipi DSI, DisplayPort, audio, sensors, GNSS, Gigabit Ethernet and more. It comes with the Open-Q 8550CS SoM, an open-frame carrier board exposing all the available IO, and accessories to fast-track product development.

Lantronix Open-Q development products are TAA and NDAA compliant, ensuring at least ten years of longevity with bill-of-materials and quality control.

Backed by more than 20 years of expertise, Lantronix’s engineering services (www.lantronix.com/engineering-services) deliver turnkey product development support for its Open-Q platforms and development kits. With engineering expertise behind more than 1500 products, its development team specialises in camera development and tuning, voice control, machine learning, mechanical and RF design, and thermal and power optimisation.

Lantronix (www.lantronix.com) is a specialist in compute and connectivity IoT that targets high-growth markets, including smart cities, enterprise and transportation.