Arduino®️ VENTUNO™ Q

Built around a dual-brain architecture that pairs a Qualcomm Dragonwing™ IQ8 processor with a dedicated STM32H5 real-time microcontroller, 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 with pre-loaded Linux 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 with expandable M.2 NVME Gen.4 port. On the MPU side, it runs standard Ubuntu (and Debian coming soon) Linux. Yet it maintains the openness, versatility, and ease of use that Arduino has become synonymous with. Docker and apt are already installed and available on the board, and you can program with VS Code, your own IDE, native Linux tools — whatever workflow your team already uses. You're never locked into it. 

Purpose-built for robotics and industrial edge AI

VENTUNO Q is engineered for systems that move, manipulate, and respond with precision. ROS 2 compatibility, real-time CAN-FD and PWM control, triple MIPI-CSI camera inputs for 360° vision, and 2.5 Gb Ethernet are all built in. Typical applications include:

• Autonomous mobile robots, drones, and pick-and-place arms – with vision-guided manipulation and Visual SLAM.
• Industrial edge AI for predictive maintenance, process automation, and energy monitoring.
• Local intelligence that processes data on-device, for anything from traffic monitoring in smart cities to responsive environments in connected homes.
• Automated quality inspection with local VLMs for on-premises defect detection.
• Education and research platforms for computer vision, generative AI, and embedded systems at any level.

One board. Three ways to build with AI

• Ready to run. A curated library of AI models, fully optimized for the IQ8 NPU via Qualcomm AI Hub, is available out of the box with no configuration required. The current library includes: Qwen 3 4B LLM; Qwen 2.5 7B and 3 4B VLM; Gemma 4 E2B and E4B; Whisper ASR; Melo and Piper TTS; YoloX small object detection; and MediaPipe gesture recognition. Additional models and Bricks are added continuously.
• Bring your own model. Upload a 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. 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.

From prototype to production

VENTUNO Q is built for industrial standards and designed to go beyond the bench. When a VENTUNO Q prototype is ready for commercial deployment, the same Dragonwing IQ8 architecture will be available as production-certified SOMs from select Dragonwing IQ8 certified partners. The software stack, AI models, and application logic developed on VENTUNO Q carry forward directly – no hardware re-architecture required.

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 the real-world data for a wide range of capabilities such as object/human detection, anomaly detection, image classification, sound recognition, and keyword spotting.

Find out more in our official documentation.

Need Help?

Check the Arduino Forum for questions about the Arduino Language or how to make your own Projects with Arduino. If you need any help with your product, please contact the official Arduino User Support through our Contact us page.

Warranty

You can find your product warranty information here.

Qualcomm branded products are products of Qualcomm Technologies, Inc. and/or its subsidiaries. 
Arduino, VENTUNO, and the Arduino logo are trademarks or registered trademarks of Arduino S.r.l.