edge ai News - CNX Software - Embedded Systems News

May 13, 2025 by Jean-Luc Aufranc (CNXSoft) - No Comments

Geniatech AP880 Edge AI embedded systems pair NXP i.MX 8M Plus SoC with 40 TOPS Kinara Ara-2 AI accelerator

Geniatech AP880 and AP880 Mini Edge AI Embedded Systems

Geniatech AP880 and AP880 Mini are two Edge AI embedded computers powered by an NXP i.MX 8M Plus AI SoC and a 40 TOPS Kinara Ara-2 AI accelerator for industrial edge applications that may benefit from AI acceleration in applications such as optical inspection. The AP880 is a single board computer design with the Ara-2 soldered on the board, while the AP880 has a modular design with a SMARC SoM, a carrier board, and an M.2 Kinara Ara-2 AI accelerator, likely Geniatech’s own AIM M2 M.2 module. Geniateach AP880 and AP880 mini specifications: SoC – NXP i.MX 8M Plus CPU – Quad-core Cortex-A53 processor @ 1.8GHz, Arm Cortex-M7 real-time core @ 800 MHz GPU – Vivante GC7000UL OpenGL ES 1.1, 2.0, 3.0, OpenCL 1.2, Vulkan support AI accelerator – 2.3 TOPS NPU VPU Encoder up to 1080p60 H.264, H.265 Decoder up to 1080p60 H.264, H.265 AI Accelerator – 40 TOPS […]

May 13, 2025 by Jean-Luc Aufranc (CNXSoft) - 1 Comment

Jupiter NX SoM – An NVIDIA Jetson Nano/NX alternative powered by SpacemiT K1 octa-core RISC-V SoC

The Jupiter NX is a system-on-module (SoM) powered by a SpacemIT octa-core 64-bit RISC-V SoC that follows NVIDIA’s Jetson Nano / Xavier NX form factor for compatibility with existing carrier boards. Designed by Shenzhen Milk-V Technology, the module supports up to 16GB LPDDR4x memory and 32GB eMMC flash, and integrates two gigabit Ethernet PHYs and a WiFi 6 & Bluetooth 5.2 wireless module. Jupiter NX specifications: SoC – SpacemIT K1/M1 CPU – 8-core X60 RISC-V (RV64GCVB) processor @ 1.6 GHz GPU – Imagination IMG BXE-2-32 @ 819 MHz with support for OpenGL ES3.2, Vulkan 1.3, OpenCL 3.0; 20 GFLOPS VPU H.265, H.264, VP8, VP9, MPEG4, MPEG2 decoder up to 4K @ 60fps H.265, H.264, VP8, VP9 encoder up to 4K @ 30fps Support simultaneous processing 1080p60 encoding + 1080p60 decoding 1080p30 H.264/H.265 encoding + 4Kp30 H.264/H.265 decoding AI performance – 2.0 TOPS (INT8) through “CPU core fusion” RVA 22 Profile […]

May 3, 2025May 3, 2025 by Jean-Luc Aufranc (CNXSoft) - 20 Comments

Walnut Pi 2B is an Allwinner T527 octa-core SBC with Raspberry Pi 5 form factor and interfaces

Walnut Pi 2B is a single board computer (SBC) powered by an Allwinner T527 octa-core Cortex-A55 SoC with a built-in 2 TOPS AI accelerator that closely follows the Raspberry Pi 5 design for compatibility with most HAT+ expansion boards and accessories. The Walnut Pi 2B SBC ships with 1GB to 4GB LPDDR4 RAM, an optional 32GB eMMC flash, and features a microSD card slot, gigabit Ethernet, four USB ports, a WiFI and Bluetooth module, MIPI DSI/CSI connectors, and the same PCIe FFC connector as found on the Raspberry Pi 5, and a 40-pin GPIO. One of the most obvious differences is that it only comes with one micro HDMI port instead of two on the Pi 5. Let’s check out the specifications to find other changes. Walnut Pi 2B specifications: SoC – Allwinner T527 CPU Octa-core Arm Cortex-A55 processor with four cores @ 1.80 GHz and four cores @ 1.42GHz […]

April 29, 2025 by Jean-Luc Aufranc (CNXSoft) - 2 Comments

Small (45x43mm) system-on-module packs Rockchip RK3562 AIoT SoC, 16GB eMMC, 2GB RAM, and PMIC

MIYR Tech MYC-YR3562 is a small 45x43mm system-on-module that packs a Rockchip RK3562 or RK3562J quad-core Cortex-A53 AIoT SoC with a 1 TOPS AI accelerator, up to 2GB LPDDR4, 16GB eMMC flash, 32 Kbit EEPROM, and a PMIC. To keep the module small while making no compromise on the number of I/Os offered, the company designed the module with 164 castellated holes (LCC – ) and 58 pads (LGA). That means the MYC-YR3562 system-on-module gets all interfaces from the RK3562(J) SoC, including Gigabit Ethernet, USB 3.0, PCIe 2.1, RGB, LVDS, and MIPI-DSI interfaces, camera inputs (MIPI-CSI), and a range of low-speed I/Os. MYC-YR3562 SoM specifications SoC – Rockchip RK3562 or RK3562J CPU RK3562 – Quad-core Arm Cortex-A53 quad-core @ 2.0 GHz RK3562J – Quad-core Arm Cortex-A53 quad-core @ 1.2 GHz (stock, overclock up to 1.8 GHz) GPU – Mali-G52-2EE with support for OpenGL ES 1.1/2.0/3.2, OpenCL 2.0, Vulkan 1.0/1.1 AI accelerator […]

April 28, 2025April 28, 2025 by Debashis Das - 1 Comment

Zalmotek RA6M1, RA8M1, and RZ/A3U SoMs follow Adafruit Feather form factor, support carrier board for robotics and industrial control

Romanian company Zalmotek has recently introduced three new SoMs, the RA6M1, RA8M1, and RZ/A3UL, and a modular carrier board designed for embedded applications such as robotics, industrial control, and edge computing. There are a few things that I find interesting about this setup. The SoM comes in Adafruit Feather form factor and, as a result, supports various Adafruit FeatherWings. The modular carrier board supports the Dynamixel motor driver module, the Particle M-SoM breakout module, Ethernet, and CAN modules. The RA6M1 Feather SoM is powered by a Renesas RA6M1 Arm Cortex-M4 CPU at up to 120 MHz, with 512 KB flash and 96 KB SRAM while the RA8M1 Feather SoM features a Renesas RA8M1 64-bit Arm Cortex-M85 running at up to 360 MHz with 128 Mbit SPI flash, the RZ/A3UL Feather SoM is based on a Renesas RZ/A3UL 64-bit Arm Cortex-A55 CPU at up to 1 GHz, with 512 Mbit OctaFlash […]

April 27, 2025April 28, 2025 by Jean-Luc Aufranc (CNXSoft) - 13 Comments

Muse Pi Pro is a feature-packed, credit card-sized SpacemIT M1 RISC-V SBC with HDMI, GbE, 4x USB, M.2 and mPCIe sockets

Muse Pi Pro is a feature-packed, credit card-sized SBC powered by the SpacemIT M1 octa-core 64-bit RISC-V AI SoC with a 2 TOPS NPU and equipped with up to 16GB LPDDR4x and 128GB eMMC flash. The single board computer features gigabit Ethernet and a WiFi 6 + Bluetooth 5.3 module for connectivity, HDMI and MIPI DSI display interfaces, two MIPI CSI interfaces, a 3.5mm audio jack, four USB 3.0 ports, an M.2 socket for an NVMe SSD or wireless module, a mini PCIE socket for WiFi and 4G LTE/5G cellular connector, and a 40-pin GPIO header for expansion. That’s quite a lot of features for such a compact board. Muse Pi Pro specifications: SoC – SpacemiT M1 (similar to SpacemIT K1, but clocked at 1.8 GHz) CPU – 8-core X60 RISC-V processor @ 1.8 GHz GPU – Imagination IMG BXE-2-32 with support for OpenGL ES3.2, Vulkan 1.2, OpenCL 3.0; 20 […]

April 25, 2025April 25, 2025 by Jean-Luc Aufranc (CNXSoft) - No Comments

IBASE EC3100 – A fanless NVIDIA Jetson Orin NX/Nano embedded computer with up to 6x GbE ports (4x with PoE)

IBASE EC3100 is a fanless edge AI computer powered by NVIDIA Jetson Orin NX or Orin Nano module with up to six GbE ports, including four with PoE for cameras, a wide 9V to 36V DC input, and a -20°C to 70°C temperature range for industrial environments. The EC3100 is equipped with a choice of modules from the Jetson Orin Nano 4GB up to the Jetson Orin NX 16GB, delivering a maximum of 157 TOPS of AI performance. The embedded system also features an HDMI 2.1 video output, audio input and output jacks, optional WiFi and/or 4G LTE/5G cellular connectivity, RS232 and CAN Bus DB9 connectors, and several digital inputs and outputs. EC3100 specifications: System-on-Module (one or the other) EC3100-NANO-4 – NVIDIA Jetson Orin Nano 4GB with 8-core Arm Cortex-A78AE CPU @ 1.5 GHz, 512-core NVIDIA Ampere architecture GPU with 16 Tensor Cores; 34 TOPS max EC3100-NANO-8 – NVIDIA Jetson […]

April 24, 2025April 25, 2025 by Jean-Luc Aufranc (CNXSoft) - 3 Comments

Espressif Systems ESP32-P4-EYE development kit looks like a camera

Espressif Systems’ ESP32-P4-EYE is a development kit designed for AI vision applications that looks like a standard camera. It combines an ESP32-P4 RISC-V microcontroller with an ESP32-C6 WiFi 6, BLE, and 802.15.4 SoC, and is equipped with a MIPI-CSI camera, a 2MP display, a microphone, and a MicroSD card to store photos or videos. There’s also a rotatory encoder to control the menu on the LCD or adjust the zoom, a fill/flash light, a few buttons, an LED, and support for either USB or battery power. The company says the ESP32-P4 camera devkit targets smart surveillance, vision-based edge AI, real-time image detection, and audio-visual IoT interfaces. ESP32-P4-EYE specifications: Microcontroller – ESP32-P4 MCU Dual-core RISC-V microcontroller @ 400 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 […]