cnx-software.com16h ago

Toradex OSM and Lino SoMs – 30×30mm NXP i.MX 93/i.MX 91 modules with solder-down or B2B connector designs

https://fed.brid.gy/r/https://www.cnx-software.com/2026/03/31/toradex-osm-and-lino-soms-30x30mm-nxp-i-mx-93-i-mx-91-modules-with-solder-down-or-b2b-connector-designs/

Toradex OSM and Lino SoMs – 30×30mm NXP i.MX 93/i.MX 91 modules with solder-down or B2B connector designs

Toradex has launched two new ultra-compact (30x30mm) System-on-Module (SoM) families: OSM and Lino, powered by NXP i.MX 91 or i.MX 93 Arm Cortex-A55 SoC for Edge industrial and IoT applications. The OSM iMX91 and OSM iMX93 variants comply with the OSM Size-S standard, featuring a 332-ball contact grid designed to be soldered to the carrier board. The Lino is a proprietary format that keeps the OSM Size-S dimensions but features two board-to-board (B2B) connectors offering more flexibility for potential replacement or future upgrades. Toradex Lino iMX91/iMX93 system-on-module Toradex Lino specifications: SoC (one or the other) NXP i.MX 93 CPU 2x Arm Cortex-A55 up to 1.7 GHz 2x Arm Cortex-M33 up to 250 MHz GPU – PXP 2D GPU with blending/composition, resize, and color space conversion NPU – Arm Ethos-U65 NPU @ 1 GHz up to 0.5 TOPS Security – EdgeLock Secure Enclave NXP i.MX 91 CPU - Single-core Arm Cortex-A55

CNX Software - Embedded Systems News
Shawn Hymel16h ago

I spoke with several ECE professors at #ECEDHA 2026 about how they’re integrating #AI into the curriculum. Everyone agrees it’s essential, but no one agrees on the best way to teach it. Check out my video interviews to learn more:

https://shawnhymel.com/3246/ai-in-engineering-education-lessons-from-ecedha-2026/?utm_source=mastodon&utm_medium=social&utm_campaign=ai_in_education

#engineer #education #edgeAI

sayzard16h ago

Ultralytics (@ultralytics)

Ultralytics가 AxeleraAI와 파트너십을 맺고, Ultralytics Python 패키지에 네이티브 내보내기(export) 통합을 추가했다. 이를 통해 YOLO 모델을 Axelera의 Metis AI Processing Units(AIPUs)로 직접 배포해 엣지 추론용 생산 환경으로 빠르게 전환할 수 있다.

https://x.com/ultralytics/status/2038964748897726533

#ultralytics #yolo #axeleraai #edgeai #deployment

Ultralytics (@ultralytics) on X

We are partnering with @AxeleraAI to introduce a native export integration built into the Ultralytics Python package. Deploy Ultralytics YOLO models directly onto Axelera's Metis AI Processing Units (AIPUs), going from a trained YOLO model to production-ready edge inference

X (formerly Twitter)
yossi18h ago
Turns out running 50 TOPS of AI inference on a module rated for -40 to +85 degrees is now a thing. Edge AI for rail, robotics, and industrial systems just got a serious hardware option. https://www.solid-run.com/blog/articles/introducing-the-p100-comx6-for-rugged-edge-ai/ #EdgeAI
Introducing The P100 COMx6 for Rugged Edge AI | SolidRun

SolidRun's new  COM Express Type 6 module family based on AMD Ryzen™ AI Embedded P100 processors, designed for developers building rugged edge AI systems.

SolidRun
cnx-software.com1d ago

Rockchip RK3506J-based Forlinx FCU1501 fanless industrial IoT gateway offers dual Ethernet, plenty of I/Os and serial interfaces

https://fed.brid.gy/r/https://www.cnx-software.com/2026/03/31/rockchip-rk3506j-based-forlinx-fcu1501-fanless-industrial-iot-gateway-offers-dual-ethernet-plenty-of-i-os-and-serial-interfaces/

Rockchip RK3506J-based Forlinx FCU1501 fanless industrial IoT gateway offers dual Ethernet, plenty of I/Os and serial interfaces

Forlinx Technology has recently introduced the FCU1501, a rugged, fanless industrial embedded computer and IoT gateway built around the Rockchip RK3506J processor with a tri-core Cortex-A7 and a single Cortex-M0 core. The system is designed for high-reliability data acquisition and protocol conversion in harsh environments. The gateway comes in two variants (basic and extended), which have similar dimensions but differ mainly in interface density. Both models feature up to 512MB DDR3 RAM, 8GB eMMC flash, dual Fast Ethernet, dual-band Wi-Fi & Bluetooth 5.0, and 4G LTE Cat 1 support. Applications include automation, rail transit, and smart manufacturing. The system supports a wide industrial temperature range (-40°C to +85°C) and meets Level 3 EMC standards. Forlinx FCU1501 Specifications: SoC – Rockchip RK3506J CPU 3x Arm Cortex-A7 cores up to 1.5 GHz Arm Cortex-M0 real-time core GPU – 2D GPU only No VPU, no NPU System Memory – 256MB or 512MB DDR3 Storage 256MB NAND

CNX Software - Embedded Systems News
ArmSoM_Official1d ago

🗣️Prior to the RK3688 launch, ArmSoM plans to develop two new products in 2026.
Something new is coming to the Sige family.🤓Sige6.
Our first @Allwinner board powered by A733.
And we want YOU to help define it.
Tell us: what would make Sige6 your next SBC?
Drop your ideas below 👇 #edgeAI #embedded

https://www.allwinnertech.com/index.php?c=product&a=index&id=139

sayzard1d ago

Alex Cheema (@alexocheema)

NVIDIA와 Apple의 협업을 암시하는 게시물로, 로컬 AI 분야에서 큰 전환점이 될 수 있는 발표가 예고됐다. 구체적인 내용은 없지만 두 기업의 결합은 온디바이스 AI, 엣지 컴퓨팅, 로컬 추론 생태계에 중요한 영향을 줄 가능성이 있다.

https://x.com/alexocheema/status/2038457155074797728

#nvidia #apple #localai #ondevice #edgeai

Alex Cheema (@alexocheema) on X

Big moment coming for local AI. @nvidia x @apple

X (formerly Twitter)
CSEM1d ago
"Think a driver is fine because they’re not yawning? Think again. DriverCheck by CSEM spots subtle physiological warning signs with contactless, privacy-first AI. See how it works: bit.ly/47lUtfQ #RoadSafety #EdgeAI"
MosChip®2d ago

The future is Agentic AI, and its here now. Are you ready for it. Listen to this snippet from our recent webinar where we discuss Agentic AI and how MosChip is enabling development of futuristic products with AgenticSky.

Here is the link to complete Webinar: https://zurl.co/tI0fv

#AgenticAI #IntelligentProducts #DigitalEngineering #MosChip #Webinar #AI #EdgeAI

cnx-software.com3d ago

M5Stack Stamp-P4 – A tiny ESP32-P4 USB-C board with optional Wi-Fi 6 and Bluetooth 5.4

https://fed.brid.gy/r/https://www.cnx-software.com/2026/03/29/m5stack-stamp-p4-a-tiny-esp32-p4-usb-c-board-with-optional-wi-fi-6-and-bluetooth-5-4/

M5Stack Stamp-P4 – A tiny ESP32-P4 USB-C board with optional Wi-Fi 6 and Bluetooth 5.4

M5Stack has just introduced the Stamp-P4, a tiny USB-C development board built around the ESP32-P4 high-performance RISC-V MCU chip, featuring 16MB of Flash and 32MB of PSRAM, and optional Wi-Fi 6 and Bluetooth 5.4 support through the ESP32-C6-MINI-1-based Stamp-AddOn C6 module. Despite its small size (29.8 x 22.0 x 4.3mm), the Stamp-P4 offers a wide range of interfaces, including a MIPI-CSI camera connector, as well as a MIPI DSI display interface, RMII Ethernet, USB 2.0 HS, and up to 44x GPIOs via 1.27mm/2.00mm pitch castellated holes and a few through holes. M5Stack Stamp-P4 specifications: SoC – Espressif Systems ESP32-P4NRW32 CPU Dual-core RISC-V microcontroller @ 360 MHz with AI instructions extension and single-precision FPU Single-RISC-V LP (Low-power) MCU core @ up to 40 MHz GPU – 2D Pixel Processing Accelerator (PPA) VPU – H.264 and JPEG codecs support Memory – 768 KB HP L2MEM, 32 KB LP SRAM, 8 KB TCM, 32MB

CNX Software - Embedded Systems News