The Raspberry Pi AI HAT+ 2 is an add-on board based on the 40 TOPS Hailo-10H AI accelerator with 8GB of dedicated on-board RAM that brings generative AI capability to Raspberry Pi 5. While it delivers similar computer vision performance as the first-generation Hailo-8-based Raspberry Pi AI HAT+, the AI HAT+ 2 also adds support for large language models (LLMs) and vision-language models (VLMs) running locally without the need for Internet access. Target applications include offline process control, secure data analysis, facilities management, and robotics. Raspberry Pi AI HAT+ 2 specifications: AI accelerator – Hailo Hailo-10H AI accelerator delivering 40 TOPS (INT4) inferencing performance Performance for computer vision models comparable to the Raspberry Pi AI HAT+ (26 TOPS) 8GB on-board RAM Host interface PCIe Gen3 x1 FPC connector to Raspberry Pi 5 40-pin GPIO header (no signal used by the Hailo-10H, it only extends the GPIO header on the Pi) […]
Radxa launches NX4 SoM with Rockchip RK3576(J) industrial SoC and NX4IO carrier board
Radxa NX4 is a 260-pin SO-DIMM SoM built around the Rockchip RK3576(J) octa-core Cortex-A72/A53 industrial SoC with a 6 TOPS NPU for edge AI workloads. It supports up to 16GB LPDDR5 memory along with optional SPI flash, eMMC 5.1 (up to 256GB), or UFS 2.0 storage (up to 1TB). Radxa has also introduced the NX4 IO carrier board for the module with an HDMI video output, two 4-lane MIPI CSI camera interfaces, four USB 3.2 Type-A ports, one USB 3.2 Type-C port, Gigabit Ethernet with optional PoE, and an M.2 M-key 2280 slot for storage, along with various I/Os. Radxa NX4 SoM specifications: SoC – Rockchip RK3576J (industrial-grade version) CPU – Octa-core CPU with 4x Cortex-A72 cores at 2.2 GHz, 4x Cortex-A53 cores at 2.0 GHz GPU – Arm Mali-G52 MC3 GPU with support for OpenGL ES 1.1, 2.0, and 3.2, OpenCL 2.0, and Vulkan 1.2 NPU – 6 TOPS […]
Rockchip RK1820/RK1828 SO-DIMM and M.2 LLM/VLM AI accelerator modules, devkits, and benchmarks
Rockchip unveiled two RK182X LLM/VLM accelerators at its developer conference last July, namely the RK1820 with 2.5GB RAM for 3B parameter models, and the RK1828 with 5GB RAM for 7B parameter models. A few months have passed, and we have a few more details since a development kit based on Rockchip RK1820/RK1828 SO-DIMM module is now available, some basic documentation has been released, and the RK1828 has been benchmarked against the Rockchip RK3588’s NPU. I was also told that M.2 modules are coming soon. Firefly RK182X 3D RAM Stacking Development Kit The “RK182X 3D RAM Stacking Development Kit” is based on the AIO-GS1N2 carrier board, previously offered with a Rockchip RK3576 or RK3588(s) and with a NVIDIA Jetson Orin Nano or NX AI module. Specifications: Main module (now only RK3588 variant for sale as part of the RK182x kit) Core-3588JD4 – Rockchip RK3588 with 6 TOPS NPU, 4GB to 32GB […]
Luckfox Aura – A Raspberry Pi-like Linux SBC powered by Rockchip RV1126B SoC with 3 TOPS NPU
The Luckfox Aura is a compact, high-performance Linux SBC built around Rockchip’s RV1126B quad-core Arm Cortex-A53 processor with a 3 TOPS NPU and designed for AI vision, multimedia processing, and edge computing Applications. Aura board supports up to 4 GB LPDDR4X memory and up to 64 GB of optional eMMC storage. Camera and display options include two 4-lane MIPI CSI camera inputs, a 4-lane MIPI DSI display interface up to 1080p60, and an AI-enhanced ISP with HDR, noise reduction, dehazing, and image correction features. The board also features Gigabit Ethernet with PoE support, onboard dual-band Wi-Fi 6 and Bluetooth 5.4/BLE, USB 3.0 OTG, four USB 2.0 host ports, a 40-pin GPIO header compatible with some Raspberry Pi HATs, audio I/O, an RTC, and a microSD card slot. Luckfox Aura specifications: SoC – Rockchip RV1126B quad-core Cortex-A53 SoC CPU – Quad-core Arm Cortex-A53 up to 1.6 GHz with 32KB L1 I-Cache and […]
Calixto Systems SL1680 OPTIMA industrial SoM and EVK features Synaptics SL1680 Edge AI processor
Calixto Systems has recently introduced the SL1680 OPTIMA, an industrial SoM built around the Synaptics SL1680 quad-core Arm Cortex-A73 SoC with a 7.9+ TOPS secure NPU for Edge AI and advanced multimedia applications. They also announced an evaluation kit (EVK) for rapid product development. The SoM offers up to 4GB LPDDR4 RAM, up to 16GB eMMC flash, an onboard Gigabit Ethernet PHY, and a wide range of interfaces through compact Hirose connectors, including dual MIPI CSI camera inputs, MIPI DSI, HDMI 2.1 Tx/Rx, PCIe 2.0, USB 3.0/2.0, SDIO, RS485, CAN-FD, and various GPIOs. The EVK further expands connectivity with MikroBUS, display and camera interfaces, and industrial I/O, making the platform suitable for AI-enabled vision systems, smart HMIs, automation equipment, and other edge computing applications. SL1680 OPTIMA SoM specifications SoC – Synaptics SL1680 (Astra Series) CPU – Quad-core Arm Cortex-A73 processor @ 2.1 GHz GPU – Imagination PowerVR Series9XE GE9920 GPU with […]
$500+ MINISFORUM MS-R1 Arm Linux AI mini PC features CIX CP8180 12-core SoC, up to 64GB LPDDR5x
MINISFORUM MS-R1 is an AI mini PC running Debian 12 Linux that’s powered by a CIX P1 (CP8180) 12-core Arm Cortex-A720/A520 processor with up to 45 TOPS of AI performance. The mini PC supports up to 64GB of soldered-on LPDDR5 memory, M.2 storage, and provides HDMI, USB-C DP, and eDP display interfaces, two 10GbE ports and WiFi 6E, a total of nine USB ports, a PCIe Gen x16 slot for a graphics card, U.2 storage, or networking expansion, a 40-pin internal GPIO header, and more. MINISFORUM MS-R101 specifications: SoC – Cix CP8180 (P1) 12-core DynamIQ processor 4x Cortex‑A720 big cores @ up to 2.6 GHz 4x Cortex‑A720 medium cores 4x Cortex‑A520 LITTLE cores Cache – 12MB shared L3 cache GPU – Arm Immortalis G720 MC10 with hardware ray-tracing support, graphics APIs: Vulkan 1.3, OpenGL ES 3.2, OpenCL 3.0 VPU Video Decoder – Up to 8Kp60 AV1, H.265, H.264, VP9, VP8, H.263, […]
Raspberry Pi CM5 industrial carrier board features dual Ethernet, optional 4G LTE/5G cellular connectivity, metal case, and more
Waveshare CM5-DUAL-ETH-4G/5G-BASE is an industrial carrier board for the Raspberry Pi CM5. It features dual Ethernet (GbE + 2.5GbE), 4G/5G options, and includes an optional metal enclosure, active cooling fan, and power supply, which is an upgrade from the previous CM5-DUAL-ETH-BASE carrier board. The board also offers two HDMI outputs supporting 4K displays, two MIPI DSI/CSI connectors for displays and/or cameras, two USB 3.2 ports, a 40-pin GPIO header, M.2 slots for both NVMe SSDs and 4G/5G modules, and PoE support via an add-on module. The board supports 7V–36V DC input, a PWM fan header, and RTC battery backup, with precise cutouts and labeled connectors for easy integration. Its main benefits over the CM5-DUAL-ETH-BASE are its more compact size and neatly arranged ports on the rear panels, while adding optional support for 4G LTE or 5G connectivity. CM5-DUAL-ETH-4G/5G-BASE specifications: Compatibility – Raspberry Pi Compute Module 5 (CM5) Storage Optional NVMe SSD […]
CRA-ready Grinn Genioboard Edge AI SBC features MediaTek Genio 510 or 700 System-on-module
Grinn has launched a credit card-sized SBC powered by its MediaTek Genio 510 or Genio 700 Cortex-A78/A55 SoM, and designed for Edge AI applications with support for the Thistle Security Platform via an on-board Infineon OPTIGA Trust M hardware security module to enable compliance with the EU’s Cybersecurity Resilience Act (CRA). With a form factor similar to the Raspberry Pi 3 model B, the Genioboard features 4GB RAM, 16GB eMMC flash, a Gigabit Ethernet port, four USB 3.0/2.0 ports, HDMI and DisplayPort video outputs, two MIPI CSI camera connectors, a 40-pin GPIO header, a USB-C debuig port, USB-C PD power, and on the bottom side, two M.2 sockets for AI accelerators and wireless modules. Genioboard specifications: System-on-Module – Grinn GenioSOM-510 or Grinn GenioSOM-700 SoC MediaTek Genio 510 (MT8370) hexa-core Arm Cortex-A78/A55 processor up to 2.2 GHz, Arm Mali-G57 MC2 GPU, 4Kp60 video decoding, 4Kp30 video encoding, up to 3.2 TOPS […]

