Arduino Ventuno Q

Arduino Ventuno Q is a dual-brain edge AI and robotics platform delivering up to 40 dense TOPS of AI acceleration alongside fast, precise control of connected hardware.

Built around a dual-brain architecture that pairs a Qualcomm Dragonwing™ IQ8 MPU with a dedicated STM32H5 real-time microcontroller, Arduino Ventuno Q doesn’t just interpret the world, it interacts with it. So you get sensing, decision, and action all on one board, on the edge, offline.

The AI brain delivers up to 40 dense TOPS of NPU acceleration for vision models, LLMs, and multi-modal AI inference. The action brain runs the Arduino Core on Zephyr RTOS, enabling sub-millisecond, deterministic control of motors, CAN-FD, PWM, and GPIO. The two communicate seamlessly via an RPC bridge: no multi-device complexity, no latency penalty, no compromise.

It’s the most powerful Arduino platform to date, offering 16 GB LPDDR5 RAM and 64 GB industrial-grade eMMC. On the MPU side, it runs standard upstream Ubuntu (and Debian coming soon) Linux. Yet it maintains the openness, versatility, and ease of use that Arduino has become synonymous for.

Docker and apt are already installed and available on the board, and it’s compatible with VS Code, PyCharm, and Jupyter - with no proprietary SDK and no lock-in.

Power supply not included.

One board - three ways to build with AI

• Ready to run. A curated library of NPU-optimized models is available out of the box via Arduino® App Lab and Qualcomm® AI Hub - no configuration required. Run Qwen and Gemma 4 LLMs and VLMs, Whisper ASR, Melo TTS and Piper, MediaPipe gesture recognition, YOLO-X object tracking, and keyword spotting. More models added continuously.
• Bring your own model (BYOM). Upload any GGUF-format model from Hugging Face or Qualcomm AI Hub directly via the llama.cpp brick in Arduino App Lab and build on it immediately – compatible with the full open-weight ecosystem, no framework migration required.
• Train your own model (TYOM). Use the integrated Edge Impulse Studio to train and quantize custom models, optimized for the Dragonwing IQ8 NPU and deployed directly into Arduino App Lab with one click.

Arduino App Lab: a unified experience (but not your only option!)

You can program Ventuno Q with the tools you are used to. Or you can enjoy the all-new Arduino App Lab, for an integrated development environment that unifies the journey across Linux and real-time OS.

Preloaded on Ventuno Q, Arduino App Lab combines Arduino Sketches, Python scripts, and containerized AI models into fully integrated applications, all managed from a single interface.

• Ready-to-use Apps and Bricks. Get started fast with Arduino Apps, self-contained examples with everything you need. Add plug-and-play features to your projects with pre-built Bricks to accelerate your ideas even more.
• Pre-loaded AI Models. With pre-loaded AI models in Arduino App Lab, you can leverage real-world data for a wide range of capabilities such as object/human detection, anomaly detection, image classification, sound recognition, and keyword spotting.

Features

• Qualcomm Dragonwing™ IQ8 (QCS8275) MPU with octa-core Arm® Cortex® CPU
• ST STM32H5F5 Arm® Cortex®-M33 real-time MCU up to 250 MHz, with 4 MB Flash and 1.5 MB RAM
• Hexagon™ Tensor AI Processor up to 40 dense TOPS, Adreno™ 623 GPU, Adreno™ VPU 623 and Qualcomm Spectra 692 ISP
• 16 GB LPDDR5 RAM and 64 GB eMMC storage
• M.2 Key M 2230 connector for non-bootable NVMe Gen 4 storage via PCIe x4
• Ubuntu Linux OS with Debian support, plus Arduino Core on Zephyr OS for the MCU
• Arduino Bridge RPC communication between the MPU and MCU
• TensorFlow Lite, ONNX Runtime and PyTorch support, with Qualcomm AI Stack, Arduino App Lab and Edge Impulse Studio integration
• Tri-band Wi-Fi® 6 (2.4/5/6 GHz), Bluetooth® 5.3 and 2.5 Gbit RJ45 Ethernet
• USB connectivity
  • 1× USB-C with host/device and power role switching, DisplayPort Alt Mode and USB Power Delivery up to 20 V
  • 2× USB 3.0 Type-A ports
  • 2× USB 3.0 interfaces via JOMEGA
• HDMI, USB-C DisplayPort Alt Mode and MIPI DSI display outputs
• 3× onboard 4-lane MIPI CSI-2 camera connectors, plus MIPI CSI0/CSI1 via JMEDIA
• MAX98091ETM+T audio codec with mono line out, mono speaker out, stereo headphone out and microphone input
• CAN-FD connectivity
  • 1× CAN-FD with ATA6563-GBQW1 PHY and onboard split termination via screw terminal
  • 3× CAN-FD without PHY via JOMEGA
  • 1× CAN-FD without PHY via UNO Shield headers
• Arduino UNO Shield-compatible headers, Raspberry Pi® HAT-compatible 40-pin JHAT and 3.3 V Qwiic connector
• Power via USB-C PD 9-20 V up to 3 A, 7-24 V 5.5×2.1 mm barrel jack up to 5 A or 7-24 V screw terminal up to 10 A
• 160 mm × 100 mm board, 25.8 mm high without the SoM heatsink and fan

Specifications

Resources

• Documentation
• Datasheet
• Arduino App Lab
• Arduino forum
• Product number: Arduino ABX00181
• EAN: 7630049205949

Package Contents

• 1x Arduino Ventuno Q

Power supply not included.