artificial intelligence (AI) News - Page 2 of 122 - CNX Software

April 21, 2025April 21, 2025 by Jean-Luc Aufranc (CNXSoft) - 2 Comments

AAEON COM-MTHC6 COM Express Type 6 Meteor Lake-H module offered with up to 45W Intel Core Ultra 9 185H processor

COM Express Type 6 Intel Core Ultra 9 185H

AAEON’s COM-MTHC6 COM Express Type 6 module is powered by the Intel Meteor Lake H processor series with a choice between the Core Ultra 9 185H, Core Ultra 7 155H, or Core Ultra 5 125H CPUs. The module supports up to 96GB DDR5 memory, features an optional BGA SSD, two MIPI CSI connectors, and two standard 220-pin B2B connectors exposing interfaces such as PEG 4.0 and PCIe Gen4, and offers a 9V to 16V power input range. AAEON claims the COM-MTHC6 is the world’s first COM Express Type 6 equipped with a 45W Meteor Lake-H CPU, such as the Core Ultra 9 185H. I can indeed see similar COM Express modules like the ADLINK cExpress-MTL only ship with 15W or 28W parts. AAEON COM-MTHC6 specifications: Meteor Lake-H SoC Intel Core Ultra 5 125H 14-core (4P+8E+2LPE) processor @ up to 4.5 GHz with 18MB cache, 7x Xe-cores Intel Arc graphics @ […]

April 16, 2025April 16, 2025 by Jean-Luc Aufranc (CNXSoft) - 2 Comments

Axelera Metis Compute Board pairs Rockchip RK3588 SoC with 214 TOPS Metis AI accelerator

Axelera Metis Computer Board is a Rockchip RK3588 mini-ITX motherboard equipped with a Metis AIPU (AI Processing Unit) capable of delivering up to 214 TOPS, 16GB LPDDR4 memory for the CPU, and 4GB LPDDR4x RAM for the AI accelerator. I first came across the Axelera Metis AIPU in M.2 and PCIe card in 2023, and I was rather impressed with the advertised 214 TOPS of AI performance promised in this form factor and relatively affordable $149 price tag considering the price-to-performance ratio. At the time, it was still hard to source the chip and module due to limited quantities, and in 2024, it became easier to evaluate the solution with the Axelera Metis PCIe Arm AI evaluation kit based on a Fireﬂy ITX-3588J mini-ITX motherboard equipped with a 214 TOPS Metis AIPU PCIe card. The new Metis Compute Board is similar, but in a more compact form factor since the […]

April 15, 2025April 22, 2025 by Sponsored Post - No Comments

Harnessing the Power of AI in Industrial Automation (Sponsored)

ModBerry 500 CM5 AI predictive maintenance

The industrial IoT (IIoT) landscape is undergoing a transformative shift, driven by advancements in artificial intelligence (AI). Among the frontrunners of this revolution is the ModBerry 500 CM5, a cutting-edge IoT platform enhanced by optional AI modules like the Hailo-8L and Geniatech AIM-M2. With the integration of these modules, the ModBerry 500 CM5 becomes a powerhouse for industrial automation, offering unparalleled performance and scalability. Let’s dive into the exciting capabilities of this innovative combination and explore the current and future applications of AI in IIoT. ModBerry 500 CM5: The Core of Smart Industrial Automation At the heart of the ModBerry 500 CM5 is the powerful Raspberry Pi Compute Module 5, featuring a quad-core Cortex-A76 processor and support for up to 16GB of RAM. When paired with AI acceleration modules like the Hailo-8L and Geniatech AIM-M2, the ModBerry 500 CM5 evolves from an IoT hub into a sophisticated AI-enabled system capable […]

April 14, 2025 by Jean-Luc Aufranc (CNXSoft) - 3 Comments

Allwinner T536 quad-core Arm Cortex-A55 & RISC-V industrial SoC supports ECC RAM, up to 3 TOP AI accelerator

Forlinx OK536-C Allwinner T536 development board

Allwinner T536 SoC features four Cortex-A55 cores, a 600 MHz RISC-V core, and a low-power RISC-V for power management, as well as support for ECC memory and an optional NPU with up to 3 TOPS of AI performance. The processor is designed to work in the industrial temperature range (-40 to +85°C) and offers plenty of I/Os, including two Gigabit Ethernet, a USB 3.1 DRD and PCIe Gen2 conbo, MIPI DSI, RGB, and LVDS display interfaces, paralle CSI and MIPI CSI camera interfaces, CAN FD, SPI, I2C, UART, several ADC, and more. The SoC is designed for interactive terminals, smart manufacturing, and other Edge AI industrial equipment. Allwinner T536 specifications: CPU cores Quad-core Arm Cortex-A55 @ up to 1.6GHz Single-core E907 RISC-V core @ up to 600 MHz Single-core E902 RISC-V MCU for low-power management GPU – G2D hardware accelerator with rotate, mixer, and scaler functions ISP Max resolution – […]

April 8, 2025April 8, 2025 by Jean-Luc Aufranc (CNXSoft) - No Comments

ADLINK COM-HPC-mMTL – A COM-HPC Mini Computer-on-Module with up to 16-core Intel Core Ultra 7 165H processor

COM-HPC-mMTL 16-core Intel Core Ultra COM-HPC Mini

ADLINK COM-HPC-mMTL is an industrial-grade COM-HPC Mini Computer-on-Module based on Intel Meteor Lake processors up to the 16-core Intel Core Ultra 7 165H featuring Intel Arc Graphics with eight Xe cores, Intel AI Boost NPU up to 8.2 TOPS, and 16x PCIE Gen4 lanes. The COM-HPC R1.2-compliant Mini size module also supports up to 64GB LPDDR5x memory soldered directly onto the board, an optional NVMe BGA SSD, and features two SATA interfaces, two 2.5 Gbps Ethernet, multiple DDI/USB4, and USB 3.0/2.0 through the standard 400-pin board-to-board connector of the 95x70mm CoM. ADLINK COM-HPC-mMTL specifications: Meteor Lake H/U SoC (one or the other) Intel Core Ultra 7 165H 16-core (6P+8E+2LPE) processor @ up to 5.0 / 3.8 / 2.5 GHz with 24MB cache, 8x Xe-cores Intel Arc graphics @ 2.35 GHz, Intel AI Boost NPU; TDP: 28W Intel Core Ultra 7 155H 16-core (6P+8E+2LPE) processor @ up to 4.8 / 3.8 […]

April 8, 2025April 8, 2025 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Aerium Lumen – An NVIDIA Jetson carrier board for drones and robots

Aerium System’s Lumen is a carrier board for NVIDIA Jetson SO-DIMM modules designed specifically for drones and robots with a range of low-profile connectors for I/Os for USB, UART, I2C, etc…, and three M.2 sockets for storage and wireless expansion. It targets the same applications as ARK Electronics’ NDAA-compliant Jetson Orin Nano/NX carrier board for drones we covered last year, but it’s not sold as a full kit and lacks NDAA-compliance. On the positive side, it is quite smaller and lighter, so it might be better suited to smaller and/or less complex drones or robots that may still benefit from Vision AI acceleration. Aerium Lumen specifications: Compatible system-on-modules NVIDIA Jetson Nano NVIDIA Jetson Orin NX NVIDIA Jetson Orin Nano NVIDIA Jetson Xavier NX NVIDIA Jetson TX2 NX Potentially other Jetson-compatible SO-DIMM modules like the Mixtile Core 3588E, but those are not listed by the company. Storage MicroSD card slot M.2 Key-B […]

April 6, 2025 by Jean-Luc Aufranc (CNXSoft) - 5 Comments

Beacon W5+ SoM – A tiny (27x15mm) Qualcomm Snapdragon W5+ System-on-Module for wearables

Beacon EmbeddedWorks’ W5+ SoM is an ultra-compact (27×15 mm) system-on-module powered by the Qualcomm Snapdragon W5+ platform for wearables with a quad-core Cortex-A53 processor, a co-processor with an Arm Cortex-M55 core and an Ethos U55 ML accelerator, and WiFi 4 and Bluetooth 5.3 connectivity. The Snapdragon W5/W5+ wearables platforms were introduced in the summer of 2022, but so far, all I could find was a $2,000 devkit (TurboX W5+) suitable for ODM/OEM manufacturers. The upcoming Beacon W5+ SoM will be one of the first hardware solutions based on the Snapdragon W5+ platform designed for commercial applications. Beacon W5+ specifications: Platform – Snapragon W5+ SW5100P-0 SoC CPU – Quad-core Cortex-A53 processor @ up to 1.7 GHz GPU – Qualcomm Adreno A702 @ up to 1,010 MHz with OpenGL ES 3.1, Vulkan 1.0, OpenCL 2.0 API support DSP – Dual Qualcomm Hexagon DSP V66K ISP – Dual ISP VPU 1080p30 Video Decode […]

April 2, 2025April 2, 2025 by Jean-Luc Aufranc (CNXSoft) - No Comments

NV8600-Nano AI Developer Kit features NVIDIA Jetson Orin Nano 8GB, quad GbE, Raspberry Pi Camera Module 2

AAEON NVIDIA Jetson Orin Nano Developer Kit Raspberry Pi Camera Modules

AAEON NV8600-Nano AI Developer Kit is based on a 67 TOPS NVIDIA Jetson Orin Nano 8GB and ships with a carrier board with quad GbE, a Raspberry Pi Camera Module 2, a fan kit (heatsink with fan), and a 60W power adapter. The carrier board is also equipped with a 256GB M.2 2280 M-key NVMe SSD, a SATA connector, HDMI 1.4 video output, two MIPI CSI connectors compatible with Raspberry Pi Camera Modules, six USB 3.2 Gen 2 Type A ports, a few serial interfaces, a 40-pin GPIO header compliant with the Jetson Orin Nano (Super) developer kit, and two more M.2 sockets for wireless/cellular expansion. NV8600-Nano AI Developer Kit specifications: NVIDIA Jetson Orin Nano 8GB Module CPU – 6-core Arm Cortex-A78AE v8.2 64-bit CPU @ 1.7 GHz with 1.5MB L2 + 4MB L3 GPU – 1,020 MHz NVIDIA Ampere architecture with 1024 CUDA cores and 32 tensor cores Video […]