machine learning News - CNX Software

June 1, 2026 by Debashis Das - No Comments

Sixfab AI HAT+ for Raspberry Pi 5 integrates DEEPX DX-M1 AI accelerator

Sixfab has launched the AI HAT+ for Raspberry Pi 5, a PCIe HAT+ based on the DEEPX DX-M1 AI accelerator, which we also found in the DEEPX DX-AIPlayer, Mini DX-M1 SoM, and ALPON X5. Unlike the M.2 module used in the ALPON X5, the AI HAT+ has the accelerator soldered directly to the board. It connects to the Pi 5 via the PCIe FFC cable and draws power from the 40-pin header. The board is also available in 13 TOPS and 25 TOPS versions and is designed to run vision AI tasks such as object detection and segmentation locally on the Pi 5. Sixfab AI HAT+ specifications: Supported SBC – Raspberry Pi 5 AI Accelerator (one or the other): DEEPX DX-M1M with up to 25 TOPS (INT8), 1 GB LPDDR4X NPU memory DEEPX DX-M1ML with up to 13 TOPS (INT8), 512 MB LPDDR4X NPU memory Host Interface – PCIe Gen […]

March 21, 2026March 21, 2026 by Debashis Das - 1 Comment

Radxa AICore DX-M1M M.2 2242 low-power AI module delivers 25 TOPS of edge AI performance for just 3W of power

Radxa AICore DX-M1M is a compact, low-power M.2 edge AI acceleration module built around the DeepX DX-M1M neural processing unit (NPU) and delivers up to 25 TOPS (INT8) of AI performance while consuming only 3W of power. Designed for industrial robot arms, autonomous mobile robots (AMR), edge servers, drones, and AIoT devices, the module delivers high-performance AI and ML capabilities without blowing the power budget. It relies on a PCIe Gen3 x2 interface and works with both x86 and Arm systems, including the Raspberry Pi 5 and Radxa ROCK SBCs. AICore DX-M1M specifications: AI Accelerator – DeepX DX-M1M neural processing unit (NPU) with up to 25 TOPS AI System Memory – 1GB LPDDR4X @ 4266 MT/s (on-chip, supports up to 8GB according to DeepX) Storage – 1Gbit QSPI NAND / NOR flash Host Interface – PCIe Gen 3.0 x4 (supports Gen 1/2/3 and x1/x2) via M.2 M + B Key connector […]

March 11, 2026 by Jean-Luc Aufranc (CNXSoft) - No Comments

Texas Instruments MSPM0G5187 and AM13Ex MCUs integrate TinyEngine NPU for Edge AI applications

Texas Instruments MSPM0G5187 and AM13Ex are two new microcontroller (MCU) families featuring the company’s TinyEngine neural processing unit (NPU) to enable low-latency, high-efficiency Edge AI/Machine Learning inference on the chips. TI claims that the TinyEngine NPU can run AI models with up to 90 times lower latency and more than 120 times lower energy utilization per inference than similar MCUs without an accelerator. The MSPM0G5187 is a general-purpose, low-power Arm Cortex-M0+ MCU, while the AM13Ex Arm Cortex-M33 microcontroller targets real-time motor control, starting with the AM13E23019 SKU. TI MSPM0G5187 general-purpose Cortex-M0+ MCU Key features and specifications: CPU – Arm Cortex-M0+ @ 80 MHz Memory – 32 KB RAM with ECC Storage – 128 KB flash with ECC, 8 KB data flash with ECC Accelerators TinyEngine NPU for AI/ML delivering up to 2.56GOPS (Giga Operations Per Second) at 80MHz MATHACL math accelerator Peripherals USB – 1x USB 2.0 (12 Mbps) Audio […]

January 7, 2026 by Jean-Luc Aufranc (CNXSoft) - 6 Comments

Nordic Semi nRF54LM20B wireless SoC integrates 128 MHz Axon NPU for Edge AI workloads

Nordic Semi nRF54ML20B Arm Cortex-M33 wireless SoC is the first nRF54L microcontroller to integrate the ultra-efficient Axon Neural Processing Unit (NPU) for edge AI workloads. The NPU is said to deliver up to 7 times faster performance and up to 8 times higher energy efficiency versus (unnamed) competing wireless solutions for tasks such as sound classification, keyword spotting, and image-based detection. The Axon NPU is also about 15x faster compared to using the Cortex-M33 CPU for the same task. Other than the NPU, the nRF54LM20B offers exactly the same features as the earlier nRF54LM20A wireless microcontroller, including up to 2 MB NVM, 512 KB RAM, a 128 MHz Arm Cortex-M33 plus RISC-V coprocessor, high-speed USB, up to 66 GPIOs, and an ultra-low-power 2.4 GHz radio supporting Bluetooth LE, Bluetooth Channel Sounding, Matter over Thread, and more as shown in the nRF54L comparison table below. Like other nRF54L wireless SoCs, the […]

December 17, 2025December 18, 2025 by Debashis Das - 2 Comments

Compact development board features a single ESP32-P4 + ESP32-C5 dual-band Wi-Fi 6 module, MIPI display and camera interfaces

WTDKP4C5 S1, ESP32 P4 and ESP32 C5 WI FI6 Development Board

Just a few months back, Wireless-Tag released the WT99P4C5-S1, which combines the ESP32-P4 with an ESP32-C5 dual-band WiFi 6 module, instead of the more commonly used ESP32-C6 wireless module found on most ESP32-P4 development boards we’ve covered. The company has now released the WTDKP4C5-S1, a more compact development board built around the WT01P4C5-S1 ESP32-P4 and ESP32-C5 core module. The board supports MIPI-CSI and MIPI-DSI through the ESP32-P4, while the SDIO-connected ESP32-C5 provides dual-band Wi-Fi 6 (2.4/5 GHz) connectivity along with BLE 5, Zigbee, Thread, and Matter. Other features include a USB 2.0 Type-C OTG port, two UART debug interfaces, two 40-pin GPIO breakouts from both chips, and various power options via USB-C, a 12V DC input, or headers. The board is suitable for LVGL-based HMIs, data acquisition, industrial control, and Edge AI applications such as IPCs and smart displays. Wireless Tag WTDKP4C5-S1 specifications: Core module – Wireless Tag WT01P4C5-S1 Main […]

December 8, 2025 by Jean-Luc Aufranc (CNXSoft) - 2 Comments

Quarky Intellio – A LEGO-compatible AI, Augmented Reality, and IoT learning platform (Crowdfunding)

Quarky Intellio is an ESP32-S3-based development kit designed as an educational platform for learning AI, Augmented Reality (AR), and IoT concepts. It is compatible with LEGO bricks and targets users aged 10 and older. The core AI-AR module features an SPI TFT display interface, a 5MP camera, a speaker and microphone, a microSD card slot for data storage, a USB-C port for programming and charging, servo and GPIO expansion ports, and a 1,000 mAh battery. The company offers a discovery kit with the core module only as well as a rover car kit, and since it’s compatible with LEGO, users can easily create their own robot. Quarky Intellio specifications: Core module – Espressif Systems ESP32-S3-WROOM-1-N16R8 SoC – ESP32-S3 CPU – Dual-core LX7 processor with up to 240MHz Memory – 512KB SRAM, 8MB PSRAM (SiP) Storage – 384KB ROM Wireless – WiFi 4 802.11b/g/n and Bluetooth 5.0 LE Storage – 16MB flash […]

December 5, 2025 by Jean-Luc Aufranc (CNXSoft) - 2 Comments

Edgi-Talk machine learning development kit features Infineon PSOC Edge E84 Edge AI SoC (Crowdfunding)

Edgi-Talk is a machine learning platform/development kit powered by an Infineon PSOC Edge E84 Arm Cortex-M55/M33 SoC featuring Arm Helium, an Arm Ethos-U55 micro NPU, and an ultra-low-power NNLite neural network accelerator, all of which enable AI/ML processing at varying power/performance levels. The devkit also comes with 128 MB PSRAM, 128MB QSPI flash, a 4.3-inch capacitive touchscreen display, two digital microphones, a speaker, WiFi 6 and Bluetooth LE 6.0 wireless connectivity, motion and environmental sensors, as well as a 40-pin Raspberry Pi header and two PMOD connectors for expansion. Edgi-Talk specifications: SoC – Infineon PSOC Edge E84 CPU Arm Cortex-M55 @ 400 MHz with FPU, MPU, Arm Helium support, 256KB i-TCM, 256KB D-TCM, and 5MB SRAM Arm Cortex-M33 @ 200 MHz with 1MB SRAM, 64KB ROM GPU – Low-power 2.5D GPU NPU – Dual architecture Arm Ethos-U55 NPU + NNLITE NPU System Memory – 128 MB PSRAM Storage 128 MB […]

November 17, 2025June 13, 2026 by Debashis Das - 1 Comment

Nuvoton NuMicro M55M1 low-power Arm Cortex-M55 MCU enables on-device AI with Ethos-U55 NPU

Nuvoton’s NuMicro M55M1 is a low-power AI MCU that combines a 220 MHz Arm Cortex-M55 CPU with a 111 GOPS Ethos-U55 NPU to run basic AI tasks on-device. It specifically targets small IoT and embedded devices that need low-power voice, audio, or simple image processing. The chip embeds 1.5 MB RAM, 2 MB flash, and supports external OctoSPI/HyperRAM memory. Connectivity options include Ethernet, USB-OTG, CAN-FD, I3C/I2C/SPI, SDIO, and an 8-bit camera interface, along with ADCs, DACs, comparators, PWM, and multiple low-power modes. It’s built for secure IoT applications with Arm TrustZone, secure boot, AES, and PSA Level 2 certification. Typical uses include voice triggers, smart sensors, simple vision nodes, small appliances, and industrial monitoring devices. Nuvoton NuMicro M55M1 specifications: CPU Core – 220 MHz Arm Cortex-M55 core Architecture – Armv8.1-M with Helium M-Profile Vector Extension (MVE) Arm TrustZone Technology DSP extensions Hardware Floating-point Unit (FPU), double-precision Cache – 16KB I-Cache […]