This year’s showcase integrates cutting-edge technologies from ADLINK’s global AI platform collaborators, including Intel, NVIDIA, MediaTek, NXP, AMD and Qualcomm Technology, Inc., highlighting ADLINK’s expertise in cross-architecture and cross-domain system integration. Furthermore, ADLINK will feature deep collaborations with AUO Display Plus for advanced display technologies and Noble Machines for the development and field deployment of humanoid robotics, presenting a comprehensive edge AI ecosystem spanning computing platforms, smart displays, and robotic applications.

Industrial‑grade computing for real‑time autonomy

The APU is responsible for running the full autonomous pipeline — from sensor ingestion to real‑time vehicle control. In DUT26, the SBC35‑ARL will enable low‑latency processing at the edge, supporting reliable autonomous operation in a demanding motorsport environment.

• Receives data from sensors including lidar, cameras, IMU and GPS, and from vehicle networks
• Runs advanced autonomy frameworks such as YOLO v11 for perception, graphSLAM for state estimation, and MPC for control
• Outputs steering and torque setpoints that are transmitted via CAN to embedded control units for actuation
• Designed to operate in real time with minimal latency — a critical requirement for autonomous racing

A partnership that advances education and innovation

Formula Student Team Delft, founded in 1999, is a multidisciplinary team of nearly 90 students from Delft University of Technology who design, build, and race a new electric race car each year. The program provides full‑cycle, hands‑on engineering experience — from research and simulation to manufacturing, integration, validation, and competition.

By partnering with the team, ADLINK supports next‑generation engineering education and accelerates innovation in autonomous mobility, giving students access to industrial‑grade edge computing for a real‑world validation platform.

Product highlight: ADLINK SBC35‑ARL

The SBC35‑ARL is ADLINK’s latest high‑performance single board computer designed for edge AI and embedded applications. In DUT26, it will serve as the central computer platform for autonomy, enabling high throughput sensor processing and deterministic control execution.

Quotes

“ADLINK is excited to enable the next generation of engineers to explore what’s possible with real‑time edge computing,” said Michael Wernaert, Sales Manager Benelux & UK at ADLINK. “Formula Student Team Delft embodies innovation, and we are proud that the SBC35‑ARL will power the autonomous intelligence of DUT26.”
 

“With ADLINK’s computing platform, we can push our autonomous stack further — processing sensor data, estimating state, and controlling the car in real time with minimal latency,” said Ziggy Beijer, Chief Embedded & Autonomous Software at Formula Student Team Delft.

• Powered by NVIDIA Blackwell architecture with NVIDIA NV FP4 AI acceleration
• High-performance GPU computing with the latest NVIDIA CUDA Cores technology, RT Cores, and Tensor Cores
• Flexible MXM 3.1 Type A/B form factors with power options from 45W–150W
• Up to 24GB high-speed GDDR7 memory for LLM and data-intensive edge AI workloads
• MIG support on EGX-MXM-BW5000 for optimized GPU resource utilization 

ADLINK Technology Inc., a global leader in edge computing, has launched the ADLINK MXM family of embedded graphics and AI accelerator modules based on NVIDIA Blackwell architecture. Offered in industry-standard MXM 3.1 Type A/B form factors, the new modules deliver scalable performance and energy efficiency for next-generation AI, robotics, and edge computing applications. They are designed to power intelligent systems that operate locally across a broad range of use cases, including robots and industrial machines, smart cameras and inspection systems, autonomous vehicles, retail infrastructure, and medical imaging and healthcare devices.

To address the wide range of deployment environments seen across various AI application demands, ADLINK’s MXM Modules powered by NVIDIA RTX PRO^TM Blackwell Generation Embedded GPUs deliver exceptional flexibility in power and performance configuration. With support for power budgets ranging from 45W to 150W, the platform enables system designers to precisely tailor performance profiles based on application priorities, whether maximizing compute capability or operating within strict thermal and energy constraints.

ADLINK’s EGX-MXM-BW2000, for example, sets a new benchmark for compact edge AI performance by delivering up to 100W of performance in a MXM 3.1 Type A form factor (82x70mm), maximizing compute density for space-constrained embedded systems. At the high end, the EGX-MXM-BW5000 features NVIDIA Multi-Instance GPU (MIG) technology that allows developers to partition GPU resources for parallel workloads, improved isolation, and more efficient resource management in advanced edge AI deployments. This robust combination enables efficient execution of large language models (LLMs), deep learning pipelines, and other data-intensive workloads directly at the edge.

“Having worked closely with customers for many years across diverse embedded applications—from medical imaging to industrial automation—we have witnessed a major shift in how AI is deployed at the edge,” says Edgar Chen, Managing Director of the Edge Visualization Business Unit at ADLINK. “As AI adoption accelerates, our customers demand higher GPU performance, larger memory capacity, and uncompromising reliability. With our MXM modules based on NVIDIA Blackwell architecture, we deliver a powerful combination of performance, efficiency, and flexibility, empowering system designers to confidently build next-generation edge AI solutions tailored to their environments.”

Built on NVIDIA Blackwell, the new MXM modules deliver breakthrough AI inference performance with FP4 precision, achieving up to 2x throughput while reducing memory bandwidth demands. The latest NVIDIA CUDA Cores, 4th-gen RT Cores, and 5th-gen Tensor Cores provide powerful GPU acceleration for AI inference, computer vision, and real-time data analytics, with peak FP32 performance ranging from 9.2 to 49.8 TFLOPS. All models include next-generation GDDR7 memory, providing high bandwidth and low latency for consistent, high-efficiency processing of complex, data-intensive workloads at the edge.

Depending on the configuration, the GDDR7 memory subsystem supports up to 24GB, memory interfaces up to 256-bit, and bandwidth reaching up to 896GB/s, significantly boosting throughput by while minimizing latency for demanding applications such as deep learning, high-resolution imaging, and advanced simulation.

All modules support up to four DisplayPort 1.4a outputs, enabling multi-display visualization for industrial, transportation, and autonomous systems. With 5-year availability and wide operating temperature ranges (Standard: 0˚C to 55˚C, Extended: -40˚C to 85˚C), the platform ensures long-term reliability, lifecycle stability, and dependable operation across both commercial and harsh industrial environments.

Product availability includes:

• EGX-MXM-BW500 – NVIDIA RTX PRO 500 Blackwell Generation Embedded GPU, MXM 3.1 Type A, 82 x 70mm
• EGX-MXM-BW2000 – NVIDIA RTX PRO 2000 Blackwell Generation Embedded GPU, MXM 3.1 Type A, 82 x 70mm
• EGX-MXM-BW4000 – NVIDIA RTX PRO 4000 Blackwell Generation Embedded GPU, MXM 3.1 Type B, 82 x 105mm
• EGX-MXM-BW5000 – NVIDIA RTX PRO 5000 Blackwell Generation Embedded GPU, MXM 3.1 Type B, 82 x 105mm

• Showcasing a unified edge AI architecture spanning embedded modules to workstation-class GPU servers for scalable industrial AI deployment.
• Featuring Intel® Core Ultra Series 3 and Xeon® 600-based systems alongside platforms powered by NVIDIA Jetson Thor for robotics and autonomous AI applications.
• Highlighting deployment-ready solutions for robotics, UAV systems, and high-performance edge AI acceleration.

ADLINK Technology Inc., a global leader in edge AI computing, will showcase its latest market-leading edge AI portfolio at Embedded World 2026 under the theme “Leading Edge AI Computing.” The showcase highlights a unified edge AI architecture spanning a full range of form factors—from COM modules and industrial motherboards to embedded systems, workstation-class GPU servers, and modular MXM and PEG GPU acceleration—designed to support scalable AI deployment in real-world environments.

As industrial AI transitions from experimentation to large-scale implementation, customers demand computing platforms that strike a balance between performance, efficiency, and long-term deployment stability. At this year’s event, ADLINK demonstrates how its portfolio supports AI from compact embedded systems to high-density server environments.

Intel-Based Edge AI Architecture — From Modules to High-Performance Systems

A key highlight is ADLINK’s expanding portfolio built on Intel® Core Ultra Series 3, delivering integrated NPU acceleration for embedded and edge AI systems. Express-PTL and COM-HPC-mPTL modules provide a compact foundation with DDR5 memory support, showcasing the performance potential of ADLINK’s embedded platforms.

Beyond hardware, ADLINK also simplifies deployment and lifecycle management. Together with AMI and Linaro, the partners are showcasing a unified embedded firmware solution for COM-based Arm systems at Embedded World 2026 that brings a true BIOS experience. This enables developers to run standard Windows or Linux distributions with full OS security updates while reducing the need for custom OS stacks.

SBC35-PTL 3.5” boards and VNX-PTL processor blades extend high-performance computing for rugged industrial applications, while the portfolio also includes complete systems, from the ultra-compact MXE-330 edge AI computer to the Twin Lake-based SP2-TWL and STC2-TWL panel PCs, delivering real-time inference, industrial visualization, and AI-enabled automation.

For higher-performance workloads, platforms based on Intel® Xeon® 600 Processors for Workstations—including the ISB-W890 server board and AXE GPU server series—deliver multi-GPU scalability and high-bandwidth processing for demanding edge AI inference.

Robotics and Autonomous AI Platforms Powered by NVIDIA Jetson Thor

ADLINK also presents platforms powered by NVIDIA Jetson Thor for robotics and autonomous systems. The DLAP-700 Series, built on NVIDIA Jetson Thor, supports real-time perception and adaptive AI processing for humanoid and collaborative industrial robotics.

Intelligent Aerial Systems Collaboration

Expanding into intelligent aerial systems, ADLINK will co-exhibit a UAV solution in collaboration with Xdynamics. The lightweight all-in-one flight control platform integrates edge computing, flight control, and RF capabilities within a compact unit, demonstrating how ADLINK’s scalable AI architecture extends from ground robotics to aerial applications.

Preparing for the Next Wave of GPU Acceleration

Looking ahead, ADLINK will preview its upcoming MXM graphics cards powered by NVIDIA Blackwell architecture within its broader platform roadmap, addressing increasing AI compute density requirements and enabling scalable GPU acceleration as next-generation technologies become commercially available.

By integrating embedded modules, edge systems, and GPU servers within a unified framework, ADLINK supports seamless scalability from compact embedded platforms to high-density server environments. The portfolio is designed to enable progressive AI deployment across industrial and mission-critical applications while maintaining operational stability and long-term scalability.

ADLINK will exhibit its edge AI portfolio at Hall 3, Booth 3-147 during Embedded World.

• Up to 180 platform TOPS for complex edge AI workloads
• Intel Arc GPU with up to 12 X^e cores simplifying system integration, especially for graphics-intensive applications
• Advanced hybrid CPU architecture combined with 128 GB DDR5, maximizing power-performance efficiency and responsiveness
• Additional industrial-temperature SKUs, featuring IBECC memory and TSN Ethernet, are suitable for mission-critical and rugged applications
• Intel Core Ultra Series 3 processors in a compact COM-HPC Mini form factor for space-constrained applications, expected in Q2 2026

New Heights in Edge AI: Significantly Faster NPU and GPU Performance

ADLINK Technology Inc., a global leader in edge AI computing solutions, proudly presents its latest Express-PTL COM modules powered by new Intel Core Ultra Series 3 processors. Express-PTL modules deliver a major performance leap with their advanced AI architecture, featuring an integrated NPU 5.0 that provides up to 50 TOPS of dedicated AI acceleration, offering significantly higher  than the previous generation, and a new  Intel Arc GPU with up to 12 X^e cores designed for efficient, streamlined operation, and delivering up to 120 TOPS for AI tasks and notably enhanced graphics rendering, supporting up to 16 CPU cores. All processing units provide up to 180 platform TOPS of AI compute, tailored not only for heavy but also for complex edge AI workloads.

High-Performance, Rugged Design for Mission-Critical Edge AI

The Express-PTL incorporates a hybrid CPU architecture with 4 high-performance (P) cores, 8 efficiency (E) cores, and 4 low-power (LPE) cores, offering improved processing efficiency and enhanced transistor performance. Complementing this high-performance computing, the module supports up to 128GB of DDR5 SO-DIMM with IBECC for low-latency, reliable memory access. Industrial-temperature SKUs are engineered to withstand extreme rugged environments with a temperature range from –40°C to 85°C. Industrial features such as TCC, TSN, In-Band ECC, extended temperature support, and FuSa/FSEDP compliance further ensure robust, mission-critical operation, making the module a versatile and dependable solution for edge AI and general embedded deployments in demanding conditions.

Versatile AI Solutions for Complex Application Workloads

The Express-PTL module delivers powerful AI performance and efficient system design across a wide range of applications:

The Express-PTL module delivers enhanced graphics performance with its integrated Intel Arc GPU, featuring a simplified design that reduces complexity and helps lower system costs. It offers enhanced graphics performance while simplifying system design and reducing overall costs, making it ideal for graphics-intensive applications such as medical imaging and infotainment.

For autonomous robots, including humanoids and quadrupeds, the module delivers fast, efficient AI computation for navigation, object recognition, and real-time decision-making. Its improved communication architecture, via a higher bandwidth PCIe interface, enables low-latency accelerator interaction, allowing robots to respond instantly while maintaining high computational efficiency.

The Express-PTL module is engineered for extreme rugged environments, supporting a –40°C to 85°C temperature range for both operating and storage conditions, making it ideal for industrial automation, outdoor systems, and edge deployments exposed to severe temperature fluctuations.

What’s Next? Intel Core Ultra Series 3 Processors Power ADLINK’s Next Wave of AI

Following the release of the Express-PTL, expect the unexpected as the COM-HPC-mPTL, SBC35-PTL, and MXE-330 expand the embedded solutions portfolio with compact and high-performance solutions designed for space-constrained, robotics, and AMR-class edge applications. The COM-HPC mPTL integrates the CPU, NPU, and GPU into a single unit with soldered-down memory, ensuring reliable operation under vibration, shock, and extreme temperatures. The efficient designs of COM-HPC-mPTL and SBC35-PTL enable high-performance AI computing and device consolidation while minimizing footprint and optimizing system power. Powered by Intel Core Ultra Series 3 processors, they deliver up to 180 platform TOPS of combined AI performance, with the integrated NPU 5.0 providing substantially higher performance and the new Intel Arc GPU faster graphics performance vs. previous generation, all within a compact form factor.

Scheduled for release in Q2 2026, the COM-HPC-mPTL will be available in a 95 mm x 70 mm module size, delivering high-performance computing for compact, rugged AI systems. Powerful and compact, the SBC35-PTL is a 3.5-inch single board computer offering energy-efficient AI computing and streamlined, space-saving device consolidation. The MXE-330 delivers system-level intelligence and speed as a versatile edge computing platform. SBC35-PTL and MXE-330 feature an Adaptive Function Module (AFM) design, enabling fully customizable I/O configurations and rapid, application-specific connectivity. Planned for launch in Q3 2026, the SBC35-PTL and MXE-330 further extend the embedded AI ecosystem with flexible, scalable, and edge AI solutions. 

ADLINK’s Express-PTL and COM-HPC-mPTL development kits, including reference carriers and full I/O support, will be available in Q2 of 2026 to help you accelerate prototyping and simplify system integration.

For more information about ADLINK Computer-on-Modules, visit adlinktech.com.

• Built on NXP i.MX 95 with up to 2 TOPS AI performance, SGeT SMARC 2.2 and OSM 1.2 compliant, for scalable, efficient edge intelligence.
• OSM-IMX95: Rugged 45 mm × 45 mm soldered-down module featuring ADLINK’s single-sided PCB design, delivering AI-powered performance for compact, power-constrained edge devices.  
• SMARC LEC-IMX95: 82 mm × 50 mm SGeT SMARC 2.2 module featuring 10 GbE, secure boot, and industrial reliability for intelligent IoT and automation.
• A wide range of options and flexible ODM customization, empowering tailored, high-performance edge solutions for diverse verticals.

ADLINK Introduces IMX95 Modules — Driving Next-Level Edge Performance

ADLINK Technology Inc., a global leader in edge computing solutions, proudly presents its latest NXP-based IMX95 modules powered by the cutting-edge NXP i.MX 95 applications processor family. Engineered to meet the demands of aerospace, automotive edge, commercial IoT, industrial, medical, and networking sectors, these modules deliver a powerful blend of performance, security, and efficiency—supported by a guaranteed 15-year product lifecycle to ensure long-term availability and stability.

At the core of the IMX95 modules is the NXP i.MX95 processor, featuring up to six-core Arm Cortex-A55 alongside Cortex-M7 and Cortex-M33 cores, providing robust multicore processing and real-time control. The integrated Neutron Neural Processing Unit (NPU), supported by the eIQ® framework, delivers up to 2 TOPS, enabling advanced AI inference, computer vision, and real-time data analytics directly at the edge—all within a low-power design optimized for sustained workloads.

Complementing this AI capability, the NXP i.MX 95 modules feature an Image Signal Processor (ISP), Video Processing Unit (VPU), and Arm® Mali graphics G310 GPU to support immersive 3D graphics and high-quality vision applications. With flexible memory configurations supporting up to 8GB LPDDR4L RAM and up to 256GB eMMC storage, plus multiple graphic output interfaces including DSI and LVDS, these modules offer extensive adaptability to varied application needs. They also include other standards found in all NXP products, ensuring broad compatibility and easy integration.

Connectivity options include USB 2.0/3.0 ports and high-speed Ethernet interfaces with Precision Time Protocol (PTP), ensuring seamless integration into diverse industrial environments.

OSM-IMX95: Ultra-Compact OSM Solution for Space-Efficient Edge Systems

The ADLINK OSM-IMX95, compliant with SGeT OSM 1.2 Size-L (45 mm x 45 mm), is a cost-effective module that delivers powerful processing and AI capabilities in a rugged, solderable form factor designed for extreme environments. Imagine a smart coffee machine that learns your preferences and brews the perfect cup every time, or portable medical analyzers providing rapid diagnostics at a patient’s bedside. Think of compact robotics navigating crowded factory floors or intelligent sensors embedded throughout smart city infrastructure, monitoring air quality and traffic flow in real time.

ADLINK adopts a single-sided PCB design for its OSM modules, following best practices to simplify manufacturing assembly, lower production and carrier board costs, and enhance reliability by minimizing thermal and mechanical stress.

Engineered for embedded edge deployments in space- and power-constrained devices, the OSM-IMX95 is available in ruggedized options that support industrial and commercial environments. It excels in smart city sensors, transportation systems, portable healthcare equipment, and compact robotics. Its soldered-down design reduces manual assembly requirements, making it ideal for high-volume production with lower labor costs. Coupled with low power consumption, integrated NPU, and industrial-grade reliability, it empowers developers to embed high-performance AI intelligence into applications where size, power efficiency, durability, and production scalability are critical.

LEC-IMX95: SMARC Module for Scalable Industrial IoT Intelligence

The ADLINK LEC-IMX95, compliant with SGeT SMARC 2.2 specifications and offered in the 82 mm x 50 mm form factor, delivers exceptional flexibility and performance for scalable edge and IoT applications. Imagine intelligent retail kiosks that recognize returning customers and personalize promotions in real time, or advanced audio systems in public venues that automatically balance clarity and volume for dynamic environments. Picture next-generation transportation systems optimizing traffic flow through AI-enabled control units, or industrial gateways managing complex sensor networks with precision and reliability.

With its optional 10GbE interface alongside the standard 2x GbE, secure boot, and real-time control, the LEC-IMX95 delivers deterministic performance across smart infrastructure, retail automation, and intelligent transport systems. Designed for scalability and easy integration, the LEC-IMX95 pairs seamlessly with cost-effective carrier boards, making it an ideal choice for both standard and low-volume production. With its long lifecycle support and industrial-grade reliability, it enables rapid prototyping and deployment of high-performance, production-ready solutions for evolving industrial and commercial environments.

Choose the Right Fit for Your Industry Innovation — Tailored-Made ODM or ADLINK-Ready Solutions

At ADLINK, we don’t just provide powerful modules — we also collaborate closely with you through advanced ODM design services to deliver tailored solutions for both OSM and SMARC platforms. Whether your project requires the modular scalability of the SMARC LEC-IMX95, enabling easy module swaps and upgrades as your needs evolve, or the ultra-compact, rugged, and cost-effective soldered-down efficiency of the OSM-IMX95, we offer the flexibility to meet your specific technical priorities.

It’s not about the industry where you’re in — it’s about choosing the right form factor for your design goals: modular expandability for future-proof adaptability, or soldered resilience for space- and cost-constrained deployments. It’s your call.

For more information about ADLINK COMs, follow these links here at adlinktech.com – LEC-IMX95 and OSM-IMX95 modules.

• AI-Ready Performance: Powered by Intel® Core Ultra with integrated NPU and optional NVIDIA® GPUs for scalable edge AI workloads.
• Modular I/O & Expansion Flexibility: Supports function modules, reserved signals, and rich I/O customization to fit diverse application needs.
• Open-Frame Industrial Design: Seamless drop-in integration into kiosks, machines, and terminals with rugged construction and wide-temperature support.
• Premium Display Quality: Features industrial-grade AUO panels with full ADLINK’s in-house customization, including optical bonding, high brightness, anti-glare, UV protection, and more.

ADLINK Technology Inc., a global leader in edge computing, proudly announces the launch of its SP2-MTL Series, a next-generation open-frame Panel PC engineered to meet the growing demand for edge AI,  in real-time, mission-critical, and data-driven industrial environments.

Powered by the latest Intel® Core Ultra H Series (Meteor Lake) processors, the SP2-MTL delivers efficient, high-performance AI inference directly at the edge. To support more advanced workloads, the system offers flexible GPU expansion via NVIDIA® MXM Type-A or PCI Express Graphics Cards, enabling scalable acceleration for applications such as machine vision, automated optical inspection (AOI), automated dispensing systems, pharmacy carousels, and smart kiosks.

Continuing the legacy of the smart panel series, the SP2-MTL retains key features that system integrators value: modular I/O expansion, reserved signal interfaces, and mechanical flexibility. Its open-frame architecture allows for seamless integration into custom enclosures, equipment, and public-facing terminals, significantly reducing design effort and time-to-market.

To ensure long-term supply and quality, the SP2-MTL features industrial-grade display panels sourced directly from AUO. Customers benefit from ADLINK’s in-house design and manufacturing capabilities, with full customization support including optical enhancements, mechanical design, and function expansions. Certified for industrial safety, EMC compliance, and shock/vibration resistance, it is purpose-built for harsh environments and mission-critical deployments.

As a long-standing technology partner of Intel® and NVIDIA®, ADLINK brings together a rich ecosystem of industrial computing, embedded GPUs, and edge AI platforms. The SP2-MTL exemplifies this synergy by delivering future-ready performance in a scalable, integration-friendly form factor. Now available, it empowers businesses to enhance efficiency, streamline operations, and unlock new possibilities across industries. Discover more at ADLINK industrial display solutions.

• Wide-Voltage Power Input: 12–28V DC-in support ensures stable operation in environments with inconsistent power supply.
• Balanced Performance: Supports Intel® Core CPUs up to 65W with PCIe 4.0, DDR4, and dual 2.5GbE.
• Industrial-Ready Design: Mini-ITX form factor with rich expansion for seamless embedded system integration.

ADLINK Technology Inc., a global leader in industrial embedded platforms and edge computing, today announced the launch of the AmITX-RL-WV, a new Mini-ITX motherboard specifically engineered for industrial embedded systems deployed in environments with variable power conditions.

Built on the Intel® H610 chipset, the AmITX-RL-WV supports 14th, 13th, and 12th Gen Intel® Core i9/i7/i5/i3 processors (up to 65W TDP) and features wide-voltage input (12–28V DC-in). This combination makes it ideal for deployment in factory automation, machine control systems, transportation, kiosks, and mobile medical carts, where power variability is a frequent challenge.

With its compact 170mm x 170mm Mini-ITX form factor, the AmITX-RL-WV delivers robust computing performance with extensive I/O and expansion capabilities, including:

• PCIe 4.0 x16 slot for add-on cards such as vision, control, or AI accelerators
• Dual 2.5GbE LAN ports for reliable industrial networking
• Up to 64GB DDR4 3200 MHz memory for stable multitasking
• 2 B/E/M-Key slots, USB 3.2 Gen2, and RS-232/422/485 COM ports for both modern and legacy device integration

“Industrial system designers often face challenges with inconsistent power supply in the field,” said HC Lin, Product Manager of Edge Computing Platforms BU at ADLINK. “The AmITX-RL-WV was developed to deliver consistent performance even in harsh electrical environments, giving our customers a dependable and versatile platform for industrial-grade embedded deployments.”

The AmITX-RL-WV expands ADLINK’s AmITX series portfolio with enhanced power resilience and mechanical compatibility for industrial applications, providing system integrators with greater design flexibility and deployment confidence.

To learn more about the AmITX-RL-WV, please visit: AmITX-RL-WV.

Summary:

• Built on AMD Ryzen Embedded 8000 Series processors for powerhouse performance.
• Deploy 8-core, 16-thread processing with AMD RDNA 3 graphics for accelerated multitasking and immersive visuals.
• Achieve up to 40 TOPS of AI inferencing performance with Zen 4, RDNA 3, and XDNA architectures for enhanced on-device intelligence.
• Support up to 96GB DDR5 with ECC for reliable data integrity and system stability.
• Enable versatile multi-display configurations (DP, eDP, HDMI, LVDS, VGA), accommodating complex visual setups with convenience.

ADLINK Technology Inc., a global leader in edge computing solutions, introduces the cExpress-R8, a COM Express® COM.0 R3.1 Type 6 Compact module powered by AMD Ryzen Embedded 8000 Series processors (8845HS, 8840U, 8645HS, 8640U). Engineered to satisfy the demands of modern smart industrial applications, it delivers an advanced integration of high-performance and energy-efficient design. This module ignites mission-critical AI capabilities, accelerated edge computing, and advanced graphics performance — bringing the confidence that comes with the reliability needed for complex, graphics-intensive workloads.

Pushing the limits of edge performance, ADLINK’s cExpress-R8 combines up to 8 “Zen 4” cores and 16 threads with integrated AMD RDNA 3 graphics and an XDNA NPU, delivering up to 40 TOPS. The cExpress-R8 fuels power-efficient AI inferencing and fast decision-making across machine vision, robotics, medical imaging, infotainment, and intelligent human-machine systems. With XDNA architecture, performance-per-watt efficiency is optimized, delivering advanced AI processing, and enabling more complex AI workloads without a significant spike in energy demands.

To guarantee real-time processing, the cExpress-R8 supports up to 96GB of DDR5 memory at 5600 MT/s, with both ECC and non-ECC options available. ECC acts as a watchdog, detecting and correcting errors in environments, where failure isn’t an option, whether in mission-critical applications, gaming, transportation, and POS (Point of Sale)/POI (Point of Information) systems.

With support for up to four displays, DP, eDP, HDMI, LVDS, and VGA interfaces. VGA and eDP outputs are available as optional features, allowing flexible visual configurations to meet edge computing demands. Combined with scalable performance, power, and graphics configurations, the cExpress-R8 is well-suited for a diverse range of edge applications. With 10-year product availability, it ensures long-term design stability and lifecycle alignment for system integrator roadmaps.

Key Features:

• AMD Ryzen Embedded 8000 Series Processors
• AMD RDNA 3 Graphics and XDNA NPU (up to 40 TOPS performance)
• Up to 96GB DDR5 5600 MT/s (ECC/non-ECC)
• Four independent display outputs

ADLINK’s cExpress development kits, including reference carriers and full I/O support, will be available in Q4 of 2025 to help you accelerate prototyping and simplify system integration.

For more information about ADLINK COMs, visit adlinktech.com and explore the cExpress-R8 module.

Summary:

• Expanded 3.5” SBC Portfolio: ADLINK introduces SBC35-MTL and SBC35-ASL, targeting both high-performance AI and rugged embedded edge applications.
• AI-Ready & Rugged Designs: SBC35-MTL features Intel® Core Ultra with integrated NPU for real-time inferencing; SBC35-ASL offers wide-temp, fanless operation with Intel® Atom® x7000RE.
• Flexible I/O with SBC-FM Modules: Both models support ADLINK’s SBC-FM expansion modules, enabling tailored I/O customization to accelerate deployment across diverse verticals.

ADLINK Technology Inc., a global leader in edge computing solutions, announces the launch of two new 3.5” single board computers: SBC35-MTL, powered by Intel® Core Ultra processors with integrated NPU, and SBC35-ASL, based on the energy-efficient Intel® Atom® x7000RE Series. These new additions to ADLINK’s SBC35 Series address a wide spectrum of edge computing demands—from compact AI workloads to industrial-grade embedded control systems.

Smarter AI in a Smaller Footprint: SBC35-MTL

The SBC35-MTL integrates Intel’s next-generation Core Ultra (Series 1) platform with Intel® AI Boost (NPU), enabling real-time AI inferencing directly at the edge. Supporting up to 96GB DDR5 memory, four independent displays, dual 2.5GbE LAN ports, and high-speed I/O such as USB 3.2, it also includes support for MIPI-CSI interfaces to connect AI cameras. SBC35-MTL is purpose-built for vision-centric applications including AMR, AI inspection, smart retail, and robotics.

Fanless Reliability for Harsh Environments: SBC35-ASL

Designed for deployment in rugged and thermally challenging environments, SBC35-ASL leverages Intel® Atom® x7835RE/x7433RE processors and offers wide temperature support (-40°C to 85°C) in a fanless configuration. The board provides rich legacy I/O including 4x COM, 8x DI/DO, and dual 1GbE LAN, making it ideal for EV chargers, outdoor kiosks, transportation systems, and industrial automation. Compatibility with ADLINK’s SBC-FM modules ensures flexible expansion while maintaining a compact footprint.

“With SBC35-MTL and SBC35-ASL, we’re extending our 3.5” SBC portfolio to serve both high-performance AI applications and rugged embedded deployments,” said Julie Huang, Product Manager at ADLINK. “These new models provide scalable performance and industrial-grade reliability, enabling our customers to rapidly deploy intelligent systems across industries. Both models also support ADLINK’s SBC-FM Adaptive Function Module platform, which enables I/O customization via FPC cable with reserved signals such as PCIe, USB, and LPC/eSPI. Developers can select from ready-to-use SBC-FM modules or design their own using our comprehensive design guide—accelerating time to market while fulfilling application-specific requirements.”

These new additions to ADLINK’s product portfolio underscore the company’s commitment to delivering compact, powerful, and customizable computing platforms for the edge. Whether deployed in intelligent automation or harsh industrial settings, the SBC35-MTL and SBC35-ASL offer the flexibility and durability to meet the evolving needs of edge AI and embedded applications worldwide.

For more product information, follow the link for the ADLINK SBC35 Series 3.5" SBCs.