CNX Software - Embedded Systems News

August 5, 2026August 5, 2026 by Debashis Das - No Comments

CtrlVibe AI Console keypad – An OpenAI Codex Micro alternative for AI workflows (Crowdfunding)

The CtrlVibe AI Console is a programmable desktop keypad designed for AI workflows. The device includes hardware controls for switching between AI tools, launching shortcuts, selecting execution modes, and recording voice input. It comes with nine programmable mechanical keys, a rotary encoder, a three-position switch, a 1.14-inch LCD, a built-in MEMS microphone, USB-C and Bluetooth connectivity, and RGB status lighting. The features and specifications of the CtrlVibe make it an alternative to Work Louder Codex Micro, a $230 mechanical macropad designed for OpenAI Codex. There is also an open-source firmware project available for the M5Stack Core2 that reproduces many of the Codex Micro’s features. While the Codex Micro is designed specifically for OpenAI Codex, the CtrlVibe AI Console can be used with multiple AI tools and uses a rotary encoder to switch between applications or models, along with a three-position hardware switch for selecting AI execution modes. CtrlVibe AI Console […]

August 4, 2026 by Jean-Luc Aufranc (CNXSoft) - No Comments

reComputer Mini J501 Edge AI computer features NVIDIA Jetson AGX Orin 64GB/32GB module, two FAKRA camera connectors

Designed for robotics applications, the reComputer Mini J501 is an Edge AI computer powered by an NVIDIA Jetson AGX Orin 64GB or 32GB module, delivering up ot 275 TOPS of AI performance. The computer features 10GbE and GbE RJ45 ports, HDMI 2.1 video output, two USB 3.2 ports, two USB 2.0 Type-C Debug/Recovery ports, a 128 M.2 NVMe SSD, two FAKRA connectors for up to eight GMSL2 cameras, and a range of other I/O connectors. reComputer Mini J501 specifications: System-on-module (one or the other) NVIDIA Jetson AGX Orin 32GB (reComputer Mini J5011 model) CPU – 8-core Arm Cortex-A78AE v8.2 64-bit processor @ 2.2 GHz with 2MB L2 + 4MB L3 cache GPU and AI NVIDIA Ampere architecture with 1792 NVIDIA CUDA cores and 56 Tensor Cores @ 930 MHz DL Accelerator – 2x NVDLA Engines @ up to 1.4 GHz Up to 200 TOPS Video Encode – 1x 4K60 | 3x 4K30 […]

Espressif ESP32-C61-MINI-1/1U Wi-Fi 6 and BLE IoT module launched for about $2

Back in 2024, Espressif Systems introduced the ESP32-C61 2.4 GHz WiFi 6 and Bluetooth LE (BLE) SoC as a low-cost variant of the ESP32-C6 without an 802.15.4 radio, a single RISC-V core, and lower-capacity SRAM (320KB vs 512KB). We had first seen the ESP32-C61 in the Wireless Tag ESP32P4C61 board introduced last June in a Kickstarter campaign last June, but Espressif has now started to sell samples of the ESP32-C61-MINI-1 and ESP32-C61-MINI-1U on AliExpress for $10.7 for a pack of 5 pieces, or around $2 per module. ESP32-C61-MINI-1/1U Module specifications: SoC – ESP32-C61 CPU core – Single-core 32-bit RISC-V microprocessor up to 160 MHz Memory 320KB SRAM Up to 8 MB PSRAM Storage 256KB ROM Up to 8 MB Quad SPI flash Wireless 2.4 GHz WiFi 1T1R Frequency: 2412 ~ 2484 MHz IEEE 802.11ax-compliant 20 MHz-only non-AP mode MCS0 ~ MCS9 Uplink and downlink OFDMA Downlink MU-MIMO Target wake time […]

August 4, 2026 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

BG2B is the lowest-power Bluetooth LE SoC from Silicon Labs so far

Silicon Labs BG2B is an ultra-low-power Arm Cortex-M33 Bluetooth 6 LE SoC with support for Channel Sounding, “Secure Vault High” security, and integrated CAN-FD and peripherals such as LED drivers and ADCs for smaller, longer-lasting IoT devices. The company claims it’s the lowest-power Bluetooth LE device in its portfolio, with 1.1 µA EM2 sleep current with RAM retention, as well as 14%-15% lower MCU active current and lower Bluetooth receive current compared to Silicon Labs BG22/BG22L. The new microcontroller targets secure ranging, asset tracking, electronic shelf labels (ESLs), smart home, industrial monitoring, and vehicle diagnostics. Silicon Labs BG2B specifications: CPU core – 32-bit Arm Cortex-M33 @ 80 MHz with DSP instructions and floating-point unit (FPU) Memory – 64 kB or 128 kB SRAM Storage – 1016 kB flash Wireless – 2.4 GHz radio Modulation Formats OQPSK DSSS 2 (G)FSK with fully configurable shaping (G)MSK Protocols Bluetooth Low Energy 6 Channel […]

August 3, 2026 by Jean-Luc Aufranc (CNXSoft) - 8 Comments

ChronoWatch X2040 – A DIY watch with off-the-shelf Raspberry Pi RP2040 round display, 3D printed case, and custom firmware

If you’ve ever wanted to build your own watch, the ChronoWatch X2040 DIY watch allows you to do it the easy way with a Waveshare RP2040-Touch-LCD-1.28 touch LCD board, a 3D-printed case, a 3.7V LiPo battery, a few screws, and firmware using the Pico C/C++ SDK. When Vishal Soni decided he wanted to build a watch, he initially planned to design a custom board, but since there are already several round LCD boards based on Raspberry Pi RP2040 MCU, he went with a Waveshare board, preferring to spend more time on firmware programming. ChronoWatch X2040 parts: Waveshare RP2040-Touch-LCD-1.28 with Raspberry Pi RP2040 MCU and 1.28-inch round touch display – About $23 to $29 on Amazon, AliExpress, or the Waveshare store. 3D printed case designed with Tinkercad 3.7V 950mAh Lipo battery (35 x 25 x 10 mm) M2 4mm screw M2 heat inserts 2-pin MX1.25 connector 1mm heat shrink sleeve tube […]

August 3, 2026August 3, 2026 by Jean-Luc Aufranc (CNXSoft) - 5 Comments

28.9M-parameter LLM runs locally on ESP32-S3 at 9 tokens/s

Slava S. (slvDev) has optimized a 28.9M-parameter LLM running locally on an ESP32-S3 development board at around 9 tokens/s while generating text, or more exactly, telling short stories. We’ve written about LLMs and ESP32 in the past, but usually the boards are used as low-power gateways to interact with online LLMs hosted on powerful servers in data centers. Such products or projects include HiWonderLLM “smart chat module”, the ESP-Claw framework, or the ESP32 Agent Dev Kit. Slava’s “esp32-ai” project is quite different, as everything runs on the ESP32-S3 microcontroller. This won’t work on all ESP32-S3 as you’ll need enough memory and storage to store the 14.9MB (4-bit) model file, and Slava tested his demo on an ~$8 ESP32-S3 board with 512KB SRAM+8MB SPRAM and 16MB flash connected to a small I2C display to display short stories. Dave Bennett already did something similar with the ESP32-LLM project, but it only relied […]

August 3, 2026August 4, 2026 by Jean-Luc Aufranc (CNXSoft) - 4 Comments

$299.99 ASUS UGen300 USB AI accelerator combines 40 TOPS Hailo-10H chip with 8GB LPDDR4

ASUS UGen300 USB AI accelerator features a 40 TOPS Hailo-10H AI processor and 8GB of LPDDR4 memory to support large language models (LLMs), vision language models (VLMs), and AI vision workloads while consuming less than 2.5 Watts. In some ways, the hardware is similar to the Raspberry Pi AI HAT+ 2, but in a slim dongle form factor (105 x 50 x 18mm) and a 10 Gbps USB 3.2 Gen1 USB-C Type-C port instead of a 5 or 8 GT/s PCIe interface. ASUS UGen300 USB AI accelerator (UGen300-USB-8G) specifications: AI accelerator – Hailo Hailo-10H delivering up to 40 TOPS (INT4) or 20 TOPS (INT8) inferencing performance Memory – 8GB LPDDR4 @ 4266 MT/s Host interface – USB 3.1 Gen2 (10 Gbps) Type-C port Power Supply – 5V via USB-C port Power Consumption – 2.5 Watts (typical) Dimensions – 105 x 50 x 18 mm Weight – 145 grams ASUS says the […]

August 3, 2026August 2, 2026 by Jean-Luc Aufranc (CNXSoft) - No Comments

Aliensense NXS is a plug-and-play GMSL 2/3 and CAN-FD sensor board for robots

Abu Dhabi-based Aliensense has introduced NXS, a compact sensor board designed to easily connect cameras and sensors to robots, without custom wiring or driver work. The board relies on a single GMSL 2/3 cable for power, video, and sensor data over up to 15 meters, extendable to 40 meters over CAN-FD (data only, reduced baudrate). The STM32G491 board is also compatible with the 2,000+ boards in MikroElektronika’s Click ecosystem thanks to a mikroBUS connector brought out through a Shuttle add-on board. This means you can use compatible sensors (via mikroBUS) in addition to, or instead of, a camera. NXS specifications: MCU – STMicro STM32G491 Cortex-M4F microcontroller with 112KB SRAM, up to 512KB flash Camera Interface – MIPI CSI-2 input, MAX96793 GMSL2/3 serializer, power-over-coax Connectivity – GMSL2/3, CAN-FD, Cyphal/serial, I2C register map Expansion – mikroBUS socket with I2C, SPI, UART, PWM, analog, INT, RST Recovery support – Signed A/B firmware, watchdog […]