BeagleBone® Black is a low-cost, high-expansion, community-supported development platform for developers and hobbyists. Users can boot Linux in under 10 seconds and get started on development in less than 5 minutes with just a single USB cable. Like its predecessors, the BeagleBone® Black is designed to address the Open Source Community, early adopters, and anyone interested in a low cost ARM® Cortex™-A8 based processor. It has been equipped with a minimum set of peripherals to allow the user to experience the power of the processor and also offers access to many of the interfaces and allows for the use of add-on boards called capes, to add many different combinations of features. A user may also develop their own board or add their own circuitry.

Compulab IOT-GATE-iMX8 is a rugged industrial edge gateway for control and monitoring deployments that require continuous 24/7 operation. It is built around the NXP i.MX8M Mini with quad-core Cortex-A53 processing at up to 1.8 GHz and a Cortex-M4 real-time co-processor, housed in a fanless aluminum enclosure. The platform supports a wide 8V to 36V DC input range, reverse-voltage protection, and extended temperature operation options. Key interfaces include dual Ethernet, three USB 2.0 ports, RS485, RS232, and CAN, with optional LTE and GNSS, optional Wi-Fi and Bluetooth, and expansion for additional serial, CAN, digital, or analog I/O. Secure boot and optional TPM support are also available.

Compulab IOT-GATE-IMX8PLUS-D1D8 is a configurable industrial gateway variant in the IOT-GATE-IMX8PLUS family, intended for deployments that prioritize local digital signal handling and control integration. It is based on NXP i.MX8M Plus processing with quad-core Cortex-A53 CPU performance, Cortex-M7 real-time control, and optional NPU capability for edge AI. The unit uses a fanless aluminum design and supports a wide 8V to 36V DC input range for cabinet and field installation. Standard interfaces include dual Ethernet, USB 3.0 and USB 2.0, RS485, RS232, and CAN bus, with optional LTE and GNSS and optional Wi-Fi and Bluetooth. Additional expansion options support tailored serial and digital I/O configurations for industrial automation workloads.

Compulab IOT-GATE-IMX8PLUS is an industrial gateway platform designed for scalable edge deployments in automation, infrastructure, and remote monitoring systems. It uses NXP i.MX8M Plus processors with quad-core Cortex-A53 compute, a Cortex-M7 real-time core, and optional NPU acceleration for AI workloads. The hardware is packaged in a fanless aluminum enclosure and supports 8V to 36V DC input for industrial installations. Connectivity includes dual Gigabit Ethernet, USB 3.0 plus USB 2.0, RS485, RS232, CAN bus, optional LTE with GNSS, and optional Wi-Fi and Bluetooth. The platform also supports configurable add-on modules for extra serial interfaces, digital I/O, analog input, mesh networking, and security features such as secure boot, watchdog, and optional TPM.

Raspberry Pi 4 Model B delivers a broad platform upgrade that moved Raspberry Pi from lightweight embedded tasks into more demanding desktop, gateway, and edge workloads. It uses the BCM2711 with a quad-core Cortex-A72 CPU and includes faster memory and I/O paths than previous generations, enabling significantly better multitasking and media performance. The board adds true Gigabit Ethernet, USB 3.0, dual-band Wi-Fi, Bluetooth, and dual display capability up to 4K-class outputs, while keeping compatibility with the established Raspberry Pi software and GPIO ecosystem. Available memory configurations and strong community support made Pi 4 a common baseline for industrial pilots, educational labs, home automation controllers, and compact Linux servers where cost, flexibility, and broad peripheral compatibility are priorities.

Raspberry Pi 5 is the latest flagship board in the Raspberry Pi line and represents a substantial generational jump in CPU, graphics, and I/O capability. It uses the Broadcom BCM2712 with a quad-core Cortex-A76 processor at up to 2.4 GHz and a VideoCore VII GPU, giving it much stronger performance for desktop-class workloads and advanced edge applications. Compared with earlier models, it adds PCIe expansion support, higher memory options up to 16 GB, and improved peripheral throughput while preserving the familiar Raspberry Pi software and GPIO ecosystem. With Gigabit Ethernet, dual-band wireless connectivity, USB, HDMI, and camera or display interfaces, it is well suited for embedded AI prototypes, automation gateways, and high-performance maker systems.

Raspberry Pi Compute Module 4 with the official IO Board provides a development and integration path for products that need Raspberry Pi 4 class compute in a module-oriented architecture. CM4 variants support multiple memory and storage combinations, including options with or without onboard eMMC and wireless connectivity, enabling design flexibility for commercial and industrial use cases. The IO Board exposes essential interfaces for bring-up and prototyping, such as networking, USB, display, camera, and GPIO-related expansion, making it practical for carrier-board design validation before custom hardware is finalized. This platform is widely used for embedded productization, OEM proof-of-concept builds, and long-lifecycle system development where module availability, interface access, and compatibility with existing Raspberry Pi software tooling are key requirements.

Seeed reComputer J3010 is a compact edge AI computer built around the NVIDIA Jetson Orin Nano 4GB module for entry-level to mid-range embedded inference tasks. It provides up to 20 TOPS in standard configuration and can reach higher performance with newer software configurations, while keeping a small deployment footprint. The system typically includes 128GB NVMe storage, Gigabit Ethernet, four USB 3.2 ports, HDMI output, dual CSI camera connectors, CAN, RTC, and 40-pin expansion. M.2 Key E and Key M support wireless and storage expansion. This combination makes J3010 suitable for machine vision, smart sensing, and robotics applications requiring practical I/O density and deployable enclosure-ready hardware.

Tinker Edge T is a single-board computer (SBC) specially designed for AI applications. It features the Google Edge TPU, a machine learning (ML) accelerator that speeds up processing efficiency, lowers power demands and makes it easier to build connected devices and intelligent applications. With this onboard ML accelerator, Tinker Edge T is capable of performing four tera-operations per second (TOPS) using only 0.5 watts per unit of computation. It is also optimised for TensorFlow Lite models, making it easy to compile and run common ML models.

Auvidea JN30B is a compact carrier board for the NVIDIA Jetson Nano production module, designed for vision-centric edge AI systems that need camera flexibility and fast local storage. The board family provides Gigabit Ethernet, optional PoE capabilities depending on variant, multiple CSI camera options, USB interfaces, and M.2 NVMe PCIe x4 support for high-speed data capture workflows. It is aimed at industrial imaging, robotics, and intelligent sensing applications where deployment size and power integration matter. The JN30B series is available in several versions with different display and I/O combinations, while keeping a similar form factor and Jetson Nano-focused architecture for straightforward product integration.

BeagleBone® AI-64, is built on a proven open source Linux approach, bringing a massive amount of computing power to the hands of developers in an easy to use single board computer. Leveraging the Texas Instruments TDA4VM SOC with dual 64-bit Arm® Cortex®-A72, C7x DSP along with deep earning, vision and multimedia accelerators, developers have access to faster analytics, more data storage options, more high speed interfaces including all the connectors you'll need on board to build applications such as Autonomous Robotics, Vision and Video Analytics, Hi-End Drones, Media Servers, and Smart Buildings.

Compulab IOT-DIN-IMX8PLUS-D1D8 is a DIN-rail industrial gateway configuration focused on applications that need stronger onboard digital I/O behavior in control and telemetry systems. It is based on the NXP i.MX8M Plus platform with quad-core Cortex-A53 processing, a Cortex-M7 real-time core, and optional NPU support for AI-enabled edge tasks. The gateway provides dual Ethernet, USB 3.0, RS485, RS232, CAN bus, and digital I/O, with optional LTE and optional Wi-Fi and Bluetooth depending on configuration. Its modular stack architecture supports up to eight expansion modules, enabling additional serial ports, digital and analog channels, mesh networking, and NVMe-based storage. The design targets long-life industrial operation with fanless cooling, secure boot, hardware watchdog, and industrial-grade environmental support.

Compulab IOT-DIN-IMX8PLUS is a modular DIN-rail gateway built for industrial cabinets and control panels that need reliable communications with expandable field interfaces. It uses NXP i.MX8M Plus processors with quad-core Cortex-A53 compute, a Cortex-M7 real-time core, and optional NPU acceleration for AI-capable workloads. The base gateway includes dual Ethernet, USB 3.0, RS485, RS232, CAN bus, and isolated digital I/O, with optional LTE and optional Wi-Fi and Bluetooth. A key capability is stacked expansion with up to eight I/O modules for extra RS485/RS232 channels, digital I/O, analog input, mesh networking, or NVMe storage. The platform is designed for 24/7 operation with fanless cooling, industrial temperature options, and secure boot plus hardware watchdog support.

Compulab CL-SOM-IMX8 is a compact system-on-module class platform based on NXP i.MX8M Mini, intended for embedded products that need balanced multimedia performance and long-term deployment stability. The processor options scale up to quad-core Cortex-A53 at 1.8 GHz and include a Cortex-M4 real-time co-processor, integrated graphics, and 1080p video encode and decode support. The module uses a very small 28 x 38 x 4 mm form factor and supports up to 4GB LPDDR4 and up to 64GB eMMC storage. Connectivity and interfaces include optional Gigabit Ethernet, optional Wi-Fi and Bluetooth, PCIe, USB 2.0, MIPI-DSI, MIPI-CSI, UART, SPI, I2C, and GPIO, making it suitable for custom industrial and edge computing carrier designs.

Compulab UCM-iMX93 is an ultra-compact system-on-module built for long-life embedded and industrial designs that need low power and strong interface flexibility. It is based on NXP i.MX93 processors with single-core or dual-core Cortex-A55 options at up to 1.7 GHz, and dual-core variants include an Arm Ethos U-65 microNPU for AI and ML workloads. A Cortex-M33 real-time core is included for control-oriented tasks. The module measures 28 x 38 x 4 mm and supports up to 2GB LPDDR4 and up to 64GB eMMC. Interfaces include Gigabit Ethernet, USB 2.0, CAN-FD, UART, SPI, I2C, MIPI-DSI, MIPI-CSI, and LVDS, with optional Wi-Fi and Bluetooth configurations.

Google's Coral Dev board with integrated TPU acceleration for fast machine learning inference on edge devices. The Dev Board is a single-board computer that's ideal when you need to perform fast machine learning (ML) inferencing in a small form factor. You can use the Dev Board to prototype your embedded system and then scale to production using the on-board Coral System-on-Module (SoM) combined with your custom PCB hardware.

FORECR DSBOARD-ORNX is an industrial carrier board platform for NVIDIA Jetson Orin Nano and Orin NX modules, designed for high-reliability edge AI deployments. The board focuses on harsh-environment operation and supports wide input voltage and industrial temperature ranges for real-world automation use. It exposes practical interfaces including Gigabit Ethernet, USB 3.1, HDMI 2.0, CAN bus, serial ports, digital I/O, CSI camera connectivity, and multiple M.2 sockets for wireless, cellular, and SSD expansion. With support for Orin NX performance tiers up to 157 TOPS, DSBOARD-ORNX fits robotics, video analytics, and autonomous systems that need compact form factor hardware with robust I/O and long-duration operational stability.

A generic 64-bit ARM platform for modern aarch64 based devices with broad compatibility.

A generic x86-64 platform supporting standard AMD64 architecture for PC-based deployments.

An extended x86-64 generic platform with additional feature support for modern 64-bit x86 architectures.

Intel's "Next Unit of Computing" (NUC) - a compact PC providing full desktop performance in a small form factor. In July 2023, Intel announced that it would no longer develop NUC mainboards and matching mini PCs. They subsequently announced that NUC products will continue to be—and since that time have been—manufactured, sold and supported by Asus under a non-exclusive license.

A Kontron i.MX8M Mini based embedded computer-on-module for industrial and IoT applications.

A compact and affordable Rockchip based single-board computer for networking and IoT applications.

The Nitrogen8M Mini SOM features NXP's i.MX8M Mini applications processors. The Nitrogen8M Mini SOM is designed for mass production use with a guaranteed 10 year lifespan, FCC Pre-scan results, and a stable supply chain. The RAM (1-4GB) and eMMC (16-128) can be modified to fit your requirements. Industrial temperature and conformal coating options are also available. No cost carrier board review is available with volume purchase.

NVIDIA Jetson AGX Orin Developer Kit is a flagship edge AI development platform for advanced robotics, computer vision, and multimodal inference pipelines. The kit includes a Jetson AGX Orin module with Ampere GPU architecture, 12-core Arm Cortex-A78AE CPU, high memory bandwidth, and support for up to 275 TOPS depending on configuration. It is designed to emulate multiple Orin module profiles, helping teams prototype and scale across the Jetson portfolio. Rich I/O, including high-speed USB, PCIe, camera interfaces, and networking, supports sensor-dense designs. With JetPack, CUDA, TensorRT, DeepStream, and Isaac integration, the platform shortens development cycles for high-complexity autonomous and edge AI systems.

NVIDIA Jetson AGX Orin 64GB Developer Kit targets top-tier embedded AI development where model size, throughput, and I/O density are all critical. The included module combines a 2048-core Ampere GPU with 64 Tensor Cores, a 12-core Arm Cortex-A78AE CPU, 64GB LPDDR5 memory, and 64GB eMMC, with configurable 15W to 60W power profiles. This platform can emulate other Orin modules and supports concurrent pipelines for vision, robotics, and generative AI workloads. Reference carrier features include PCIe expansion, M.2 slots, CSI camera connectivity, USB 3.x, and high-speed networking. It is well suited for teams building production-grade autonomous systems that require robust prototyping headroom.

NVIDIA Jetson Nano 2GB Developer Kit is an entry-level edge AI platform aimed at education, prototyping, and lightweight vision inference workloads. It uses the same Jetson Nano compute architecture with a 128-core Maxwell GPU and quad-core Cortex-A57 CPU, but in a reduced-memory configuration to lower cost and simplify adoption. The board supports common maker and embedded interfaces such as CSI camera input, USB, Ethernet, HDMI display output, and general expansion headers depending on carrier implementation. It is widely used for introductory robotics, object detection demos, and proof-of-concept deployments where software compatibility with JetPack and CUDA-based tooling is important but high TOPS-class performance is not required.

NVIDIA Jetson Nano eMMC module variants are designed for embedded AI products that need fixed onboard storage and production-friendly integration. The platform combines a 128-core Maxwell GPU, quad-core Arm Cortex-A57 CPU, and software support through JetPack, CUDA, and TensorRT for compact inference workloads. Compared with removable-media developer setups, eMMC-based configurations are better aligned to controlled manufacturing and deployment processes in robotics, smart cameras, and industrial gateways. Typical systems built on Nano eMMC expose USB, Ethernet, CSI camera interfaces, and display output through carrier boards, providing enough I/O for practical edge analytics while maintaining low power draw and a broad ecosystem of pretrained models and reference applications.

NVIDIA Jetson Nano is a compact AI computing platform for low-power edge applications such as smart vision nodes, educational robotics, and embedded inference prototypes. It is based on a quad-core Arm Cortex-A57 CPU and 128-core Maxwell GPU, offering enough compute for entry-level neural network workloads while maintaining accessible power and cost profiles. The module and associated developer platforms are supported by JetPack software, enabling CUDA acceleration, multimedia processing, and common AI framework integration. Depending on carrier design, Jetson Nano systems can provide Ethernet, USB, CSI camera inputs, and HDMI output, making the platform a practical starting point for teams that need proven NVIDIA tooling and ecosystem compatibility in small-footprint deployments.

Jetson Orin Nano developer configurations with NVMe storage target modern entry-level edge AI workloads that exceed classic Jetson Nano performance. The Orin Nano platform uses an Ampere GPU architecture with Tensor Cores and Arm Cortex-A78AE CPU cores, delivering substantially higher TOPS throughput for real-time vision and multi-model inference. NVMe-backed setups improve local dataset handling, model loading, and logging performance compared with slower removable media approaches, which is valuable for iterative development and field validation. Typical interfaces include Ethernet, USB, HDMI, and CSI camera support through development carriers. This makes Orin Nano NVMe systems a strong option for compact robotics, intelligent video analytics, and production-oriented edge AI prototyping.

Jetson Orin NX and Xavier NX compatible developer platforms provide a flexible path for teams that need to balance software continuity with scalable AI performance. Xavier NX offers Volta-based acceleration for mature deployments, while Orin NX introduces Ampere architecture and higher TOPS capacity for more demanding perception and analytics pipelines. These kits are commonly used in robotics and industrial vision projects where camera throughput, low-latency inferencing, and reliable peripheral expansion are required. Typical carrier features include Ethernet, USB, CSI camera connectors, display outputs, and optional wireless modules. By supporting both module families, the platform enables staged upgrades from existing Xavier NX designs to Orin NX performance tiers with reduced integration risk.

The Jetson TX2 NX developer kit is an entry-to-mid tier embedded AI platform that combines TX2-class Pascal GPU acceleration with a compact Jetson NX form factor. It provides up to 1.33 TFLOPS AI compute using a 256-core Pascal GPU and dual Denver 2 plus quad Cortex-A57 CPU architecture, with 4GB LPDDR4 memory and 16GB eMMC on the module. Typical kits add high-speed interfaces such as Gigabit Ethernet, USB 3.x, CSI camera support, HDMI or DisplayPort output, and optional NVMe expansion through carrier boards. The platform is suited for prototyping and deployment of smart cameras, automation nodes, and compact robotics products requiring predictable low-power edge inference.

NVIDIA Jetson TX2 is an embedded AI module built for power-constrained edge systems that still require substantial GPU acceleration. It combines a 256-core Pascal GPU with dual Denver 2 and quad Cortex-A57 CPU cores, 8GB LPDDR4 memory, and 32GB eMMC storage, operating in 7.5W and 15W power profiles. The platform is intended for robotics, machine vision, and smart automation products where compact size and efficient inferencing are critical. Jetson TX2 also benefits from the JetPack software ecosystem, including CUDA and accelerated AI libraries, which simplifies deployment of deep learning and computer vision pipelines on production hardware with long lifecycle expectations in embedded environments.

NVIDIA Jetson Xavier is a high-performance embedded AI platform family designed for autonomous machines that need real-time perception, planning, and control. Xavier modules combine NVIDIA Volta GPU acceleration with Carmel Arm v8.2 CPU cores, delivering significantly higher AI throughput than earlier Jetson generations while preserving embedded power envelopes. Depending on model, the family supports advanced workloads such as visual SLAM, sensor fusion, multi-camera analytics, and robotics autonomy stacks. The platform is tightly integrated with JetPack software, CUDA, TensorRT, and vision frameworks, making it practical for production edge deployments in robotics, industrial inspection, and intelligent transportation where deterministic performance and broad ecosystem support are required.

Jetson Xavier NX eMMC-based developer configurations provide the same Xavier NX compute architecture while using onboard eMMC-oriented module variants for embedded product workflows. The platform combines a 6-core Carmel CPU with a 384-core Volta GPU and Tensor Cores, enabling around 21 TOPS class inference performance in compact, power-efficient designs. Typical carrier implementations include CSI camera interfaces, USB 3.x, Gigabit Ethernet, and display outputs for rapid AI prototyping and deployment. This setup is well suited for vision analytics, autonomous machines, and smart industrial endpoints where deterministic storage behavior and integrated module-based deployment are preferred over removable media development paths in production.

NVIDIA Jetson Xavier NX Developer Kit delivers Xavier-class AI performance in a compact edge form factor for robotics and intelligent vision systems. It is built around a Xavier NX module with a 6-core Carmel CPU and 384-core Volta GPU with Tensor Cores, providing up to 21 TOPS for inference-heavy workloads. Typical developer kit and carrier-board implementations expose CSI camera lanes, USB 3.x, Gigabit Ethernet, HDMI or DisplayPort, and M.2 expansion for storage or wireless modules. The platform supports the JetPack software stack with CUDA, TensorRT, and DeepStream, enabling rapid deployment of object detection, segmentation, and multi-sensor analytics in power-efficient embedded products.

The i.MX 8M Mini EVKB provides a platform for comprehensive evaluation of the i.MX 8M Mini and i.MX 8M Mini Lite applications processors. It delivers high performance with power efficiency, multimedia interfaces, and Wi-Fi/Bluetooth for connectivity out-of-the box. The two-board solution consists of a compact 2”x2.7”compute module and a larger base board that brings out the broad connectivity that is needed for product evaluation. The compute module is a proven, compact reference to accelerate your own design.

owa5X is a rugged wireless embedded Linux computer designed for off-highway and vehicular edge deployments that require reliable communications, rich field I/O, and high durability. It uses an NXP i.MX8M Plus quad-core Cortex-A53 processor at up to 1.6 GHz, with optional NPU acceleration for AI workloads and configurable memory or storage options including eMMC and microSD. The platform supports LTE, Wi-Fi, Bluetooth, GNSS, CAN-FD, RS-232 and RS-485 serial interfaces, Ethernet, USB, plus digital and analog I/O for sensor and actuator integration. It includes industrial security features such as secure boot and TPM 2.0, and is built in an IP6K9K enclosure for harsh environments and long-term field operation.

At the heart of the phyBOARD®-AM62x is the industrial phyCORE-AM 62x, a versatile module for a wide range of embedded applications with scalability for features as well as cost optimization. This rich multimedia insert ready single-board computer (SBC) is equipped with Ethernet, CAN, UART, I2C, SPI , dual display, MIPI CSI-2 camera, and audio. As a NextGen Sitara Application the phyBOARD®-AM62x offers a modern FTDI interface allows software download and debugging, perfect for development. Paired alongside the a M.2 connectorized Wifi and bluetooth module extension making this product perfect for IoT devices.

The original Raspberry Pi Model B introduced a low-cost Linux single-board computer platform that made embedded computing broadly accessible to education, hobbyist, and prototyping communities. It is built on the Broadcom BCM2835 with a 700 MHz ARM11 processor and VideoCore IV graphics, and it established the basic Raspberry Pi architecture adopted and expanded in later generations. The board provided core interfaces for practical projects, including Ethernet networking, USB connectivity, HDMI display output, and GPIO expansion pins for hardware interaction. It was designed to support software learning and hands-on programming while remaining affordable and compact. This model became the foundation for the broader Raspberry Pi ecosystem now used across education, industrial control, IoT experimentation, and custom device development.

Raspberry Pi 2 Model B marked a major performance upgrade over first-generation boards by introducing a quad-core Cortex-A7 processor and 1 GB of RAM in the same familiar form factor. This generation kept the practical interface mix expected by existing users, including Ethernet, USB, HDMI, and GPIO, while delivering a much more responsive Linux experience for development and education workloads. It provided enough additional compute headroom for richer desktop use, larger software stacks, and more capable edge prototypes without requiring changes to many existing accessories and workflows. Raspberry Pi 2 helped bridge early hobbyist use into broader embedded and light industrial experimentation, making it a widely adopted stepping stone between the original ARM11-based models and later 64-bit Raspberry Pi generations.

Raspberry Pi 3 Model B combines a 64-bit quad-core Cortex-A53 CPU with integrated Wi-Fi and Bluetooth, creating a balanced single-board platform for connected embedded projects. It retains core interfaces expected across the ecosystem, including Ethernet, USB, HDMI, CSI, DSI, and GPIO, which makes it suitable for sensor integration, edge data collection, and general Linux application development. Compared with older Raspberry Pi boards, the integrated wireless capabilities simplify enclosure design and lower total system complexity for IoT and automation use cases. Pi 3 is widely deployed in education, maker projects, and lightweight industrial prototypes because it offers stable software support, broad accessory compatibility, and enough compute for scripting, control logic, and moderate networking workloads in compact, low-cost systems.

Raspberry Pi 3 Model B brought integrated wireless connectivity into the mainstream Raspberry Pi platform while introducing a 64-bit quad-core Cortex-A53 processor architecture. This generation combined useful edge interfaces, including Ethernet, USB, CSI, DSI, HDMI, and GPIO, with onboard Wi-Fi and Bluetooth to reduce external adapter requirements in compact deployments. It became a practical option for connected IoT gateways, educational computing, robotics prototypes, and home automation systems that needed Linux flexibility with better performance than earlier boards. Because it remained compatible with many existing accessories and software workflows, Pi 3 helped accelerate adoption in both hobbyist and semi-industrial environments. It continues to be used where low power, modest compute, and integrated wireless communication are important design constraints.

Raspberry Pi Zero 2 W brings a significant compute upgrade to the ultra-compact Zero family by moving to a quad-core 64-bit Cortex-A53 class architecture while maintaining a small footprint and low power profile. It is designed for embedded and IoT deployments where physical size, cost, and wireless connectivity matter more than high-end throughput. The board integrates Wi-Fi and Bluetooth, and supports common Raspberry Pi workflows for Linux-based control, lightweight edge processing, and remote device management. With USB, mini HDMI, and GPIO availability in a compact format, Zero 2 W is frequently used in portable devices, smart sensors, camera nodes, and automation endpoints. It offers a practical balance between performance and simplicity for space-limited connected systems.

RevPi Connect 4 is a modular industrial PC and edge gateway platform based on Raspberry Pi Compute Module 4, designed for demanding IIoT, automation, and control applications. It combines a Broadcom BCM2711 quad-core Cortex-A72 processor with configurable memory and eMMC storage options, providing a substantial performance uplift over earlier generations. Integrated industrial interfaces make it suitable for both OT and IT integration, including dual Ethernet, USB 3.2, micro-HDMI, RS-485, and configurable digital or relay I/O functions. The device can be expanded with compatible RevPi I/O modules through PiBridge, enabling flexible adaptation to project requirements. With industrial design principles and Linux openness, RevPi Connect 4 is suitable for controller, gateway, and data acquisition roles.

RevPi Connect S is a compact industrial control and gateway variant in the Revolution Pi Connect family, optimized for DIN-rail deployments that need reliable OT connectivity and open Linux flexibility. It is based on Raspberry Pi Compute Module 4S with Broadcom BCM2711 quad-core Cortex-A72 processing and configurable eMMC options, providing a practical balance of performance and robustness. The platform is intended for IIoT integration tasks such as machine data acquisition, protocol bridging, and edge application hosting. Integrated interfaces include Ethernet, USB, micro-HDMI, and RS-485, while modular expansion with compatible RevPi components enables project-specific adaptation. Its open system approach and industrial-grade design make it suitable for long-term automation and remote monitoring deployments.

An NXP i.MX8M Plus based industrial edge gateway with multiple connectivity options for IoT and edge computing.

Radxa Rock Pi 4B with powerful RK3399 SoC, suitable for high-performance computing and multimedia applications.

Seeed J4012-based Jetson platform configurations target higher-end edge AI deployments using NVIDIA Jetson Orin NX modules with strong inference throughput and flexible expansion. The platform is designed for computer vision, robotics, and industrial intelligence workloads that need camera input, high-speed peripheral connectivity, and stable embedded operation. Typical interface coverage includes CSI camera lanes, Gigabit Ethernet, USB, HDMI or DisplayPort output, and M.2 expansion for storage and communications. With Orin NX compute options offering substantially higher TOPS than entry-level modules, J4012-class systems provide a practical bridge between development and production, especially for applications requiring multi-stream analytics, low-latency decision pipelines, and rugged deployment packaging.

The DART-MX8M offers an ideal solution for embedded systems that require high-end multimedia applications in a small form factor, as well as portable and battery operated products. This tiny 30mm x 55mm ARM SoM / CoM is based on NXP i.MX8M Quad 1.5GHz ARM Cortex-A53 plus 266MHz Cortex-M4. The high multimedia performance spec encompasses 4K video HEVC/H265/H264/VP9 decode with HDR, high-quality audio, 4K display support, 2D/3D graphics acceleration, with a variety of interfaces and connectivity options: Certified dual-band Wi-Fi 802.11 ac/a/b/g/n, BT 5.2/BLE, GbE, dual USB3, dual PCIe and UART. The DART-MX8M SoM provides a pin2pin scalable option to the DART-MX8M-MINI based on NXP’s i.MX8M Mini 1.8GHz Single/Quad-core Cortex-A53 plus 400MHz Cortex-M4 real-time processor with integrated video encode and decode acceleration, lower power consumption and higher CPU performance. For the complete Pin2Pin product list, check the Variscite Pin2Pin System on Module families.

Packed in a tiny 55 x 30 mm package, the DART-MX8M-MINI System on Module / Computer on Module based on NXP's i.MX8M Mini with up to 1.8GHz Quad-core ARM Cortex-A53™ plus 400MHz Cortex-M4™ real-time processor, leveraging advanced low power silicon process technology to provide optimized power consumption while maintaining a high-performance bar. The DART-MX8M-MINI provides a pin2pin scalable option to the DART-MX8M-PLUS and the DART-MX8M with integrated 1080p video encode and decode acceleration support, 2D and 3D graphics, HQ audio and a wide range of connectivity options such as certified Wi-Fi 6 dual-band 802.11ax/ac/a/b/g/n with BT/BLE 5.4 and optional 802.15.4, Ethernet and USB.

Packed in a tiny 55 x 30 mm package, the DART-MX8M-MINI System on Module / Computer on Module based on NXP's i.MX8M Mini with up to 1.8GHz Quad-core ARM Cortex-A53™ plus 400MHz Cortex-M4™ real-time processor, leveraging advanced low power silicon process technology to provide optimized power consumption while maintaining a high-performance bar. The DART-MX8M-MINI provides a pin2pin scalable option to the DART-MX8M-PLUS and the DART-MX8M with integrated 1080p video encode and decode acceleration support, 2D and 3D graphics, HQ audio and a wide range of connectivity options such as certified Wi-Fi 6 dual-band 802.11ax/ac/a/b/g/n with BT/BLE 5.4 and optional 802.15.4, Ethernet and USB.