Product Description
The Arduino® VENTUNO™ Q is an edge AI computer built for applications that require both local AI inference and deterministic real-time control in a single platform. Its dual-brain concept pairs Linux-based compute for AI workloads (vision, audio, local models) with a dedicated MCU path for time-critical actuation such as motor control.
Dual-brain architecture: Linux compute plus real-time control
The platform combines Qualcomm Dragonwing™ IQ8 (QCS8275) with an STM32H5F5 microcontroller. Linux applications can be coupled to the Arduino/MCU domain via RPC-style mechanisms to build integrated sense–decide–actuate pipelines without separate controller boards.
AI performance: up to 40 Dense TOPS for vision and local models
For AI inference, the specification lists a Hexagon Tensor AI Processor (NPU) delivering up to 40 Dense TOPS. This targets edge workloads such as object detection, classification, segmentation, tracking, audio intelligence (keyword spotting, speech pipelines) and local model execution in constrained environments.
Connectivity and I/O for prototyping and integration
VENTUNO™ Q integrates key interfaces on-board: 2.5Gbit Ethernet, Wi-Fi® 6 and Bluetooth® 5.3, plus high-speed I/O including USB-C (host/device role switching, DP Alt Mode video output), USB 3.0 Type-A and HDMI. For camera/display connectivity, MIPI CSI and MIPI DSI are listed via carrier headers.
Storage and expansion: eMMC plus NVMe Gen4
On-board storage is listed as 64GB eMMC. Storage can be expanded via an M.2 slot for NVMe Gen4 external drives—useful for datasets, logs, containers and models.
Ecosystem compatibility: UNO shields, carriers, Modulino and RPi HATs
For fast reuse of peripherals, VENTUNO™ Q combines multiple ecosystems: Arduino® UNO shield headers, UNO carrier headers (JMEDIA/JMISC), Qwiic for Modulino® nodes, and a 40-pin connector for Raspberry Pi® HATs. An 8×13 LED matrix and RGB LEDs are included for feedback and UI prototyping.
Software stack: Linux + Zephyr, containers and Arduino App Lab
The MPU OS list includes Linux Debian and Canonical Ubuntu. The real-time domain is specified as Arduino Core on Zephyr OS. The platform also lists Docker and Docker Compose support and Arduino App Lab to unify sketch, Python and AI workflows.
For project inquiries, volume planning, or integration guidance, please use our contact page, call 089 895050, or email store-ate@atxx.de.
- Product: Arduino® VENTUNO™ Q
- MPU / SoC: Qualcomm Dragonwing™ IQ8 (QCS8275) on a BGA SoM
- CPU: Octa-core Arm® Cortex® (as specified)
- GPU/VPU: Adreno (A623 @ 877 MHz, as specified)
- NPU: Hexagon Tensor AI Processor – up to 40 Dense TOPS
- ISP: Qualcomm Spectra 692 ISP
- MCU: STM32H5F5 (Arm® Cortex®-M33 @ 250 MHz), 4MB flash, 1.5MB RAM
- RAM: 16GB LPDDR5 (2×8GB)
- Storage: 64GB eMMC; M.2 connector for NVMe Gen4 external storage
- USB: 1× USB-C (host/device, power role switch, video output), 2× USB 3.0 Type-A, 2× USB 3.0 on JOMEGA
- Connectivity: Wi-Fi® 6 (2.4/5/6 GHz), Bluetooth® 5.3, 1× 2.5Gbit RJ45 Ethernet
- CAN: 1× CAN-FD PHY on screw terminal; 3× CAN-FD (no PHY) on JOMEGA; 1× CAN-FD (no PHY) on UNO shield headers
- Camera: USB camera support; 3× MIPI CSI connectors (muxed) with 2× MIPI CSI on JMEDIA header
- Video: 1× HDMI muxed with MIPI DSI on JMEDIA; DP Alt Mode video output via USB-C; MIPI DSI pins on JMEDIA
- Interfaces: I2C/I3C, SPI, PWM, UART, PSSI, GPIO, JTAG, ADC
- Audio: 2× microphone in / headphone out / ear out / line out on JMISC header
- Power supply: USB-C 5VDC max 3A; 12–24VDC power jack (5.5×2.1 mm); 7–24VDC screw terminal; 7–24V on JOMEGA
- Dimensions: 160 × 100 × 25.8 mm
- Temperature range: Commercial: −10°C to +60°C
- Software: Linux Debian/Ubuntu; Arduino Core on Zephyr OS; Docker/Docker Compose; Arduino App Lab
- SKU (Sales Brief): ABX00181
- Barcode (Sales Brief): 7630049205949
- Vision-guided robotics: pick-and-place, sorting, manipulators (AI perception + deterministic actuation)
- AMRs & mobile robotics: Visual SLAM/path planning on Linux with MCU-driven real-time motor/obstacle control
- Industrial inspection: defect detection, quality inspection, real-time anomaly detection with fast response I/O
- Predictive maintenance: sensor anomaly detection and local decision logic
- Smart city / security: edge traffic monitoring, site safety/security monitoring (use-case oriented)
- Vision-based inventory monitoring: shelf/stock monitoring and alerting
- AI assistants & smart environments: offline voice, local reasoning hubs, multimodal automation
- Education & research: AI learning kits and rapid research prototyping (ROS 2 context as referenced)
FAQ
What does “dual-brain architecture” mean on Arduino® VENTUNO™ Q?
VENTUNO Q pairs Linux compute for AI workloads with a dedicated STM32H5 MCU path for deterministic real-time control, enabling integrated sense–decide–actuate pipelines on one platform.
How much AI performance is specified?
The specification lists a Hexagon Tensor AI Processor (NPU) delivering up to 40 Dense TOPS on the Qualcomm Dragonwing™ IQ8 (QCS8275) platform.
What storage options are supported?
Listed are 64GB eMMC on-board and an M.2 slot for NVMe Gen4 external storage. Memory is specified as 16GB LPDDR5 (2×8GB).
Which network and wireless interfaces are integrated?
The specs include Wi-Fi® 6 (2.4/5/6 GHz), Bluetooth® 5.3, and 2.5Gbit Ethernet (RJ45).
How are cameras and displays connected?
The platform lists USB camera support and MIPI CSI connectivity (connectors and JMEDIA header). Video includes HDMI (muxed with MIPI DSI via JMEDIA) and video output via USB-C (DP Alt Mode).
Is VENTUNO Q compatible with Arduino UNO shields and Raspberry Pi® HATs?
Yes. The board lists UNO shield headers and a 40-pin connector (RPi HAT form factor). It also includes Qwiic/Modulino® and UNO carrier headers (JMEDIA/JMISC).
Which operating systems and development paths are listed?
The product was launched on the market by the manufacturer before December 13, 2024
and was offered for sale by us before December 13, 2024.
The product conforms to Directive 2001/95/EC.