Arduino has opened pre-orders for VENTUNO Q, described as its most advanced edge AI platform to date, powered by Qualcomm's Dragonwing IQ8 Series processor. Now available worldwide, VENTUNO Q was built specifically with developers in mind, emphasizing familiarity and ease of use through supporting software, accessories, and extensions.
The board's core innovation is what Arduino calls a "dual-brain" architecture, pairing a Dragonwing IQ-8275 processor, delivering up to 40 dense TOPS (trillion operations per second) of AI performance, along with 16 GB of LPDDR5 memory, 64 GB of eMMC storage, and expandable M.2 NVMe storage, with a separate STM32H5 real-time microcontroller. This combination lets machines built on the platform not just interpret sensor data using AI, but also respond to it physically, immediately, and entirely offline, without needing a network connection.
The board ships with Ubuntu preloaded, distributed by Canonical, and Arduino's App Lab software offers a quick way to get a working prototype running. Developers can also use VS Code, standard Linux development tools, or their own existing workflows if preferred. On the microcontroller side, Arduino's core software running on the Zephyr real-time operating system is designed to provide precise, real-time control over motors, CAN-FD communication, pulse-width modulation (PWM), and high-speed general-purpose input/output (GPIO) pins.
Fabio Violante, vice president and general manager of Arduino at Qualcomm Technologies, said that for 21 years, Arduino has taken complex technology and made it simple enough for people to build something real, and that VENTUNO Q applies that same mission to what he called the most consequential technology of the current era. He said the platform gives developers, educators, and innovators the tools to build machines that don't just think, but act.
Through Arduino App Lab, developers can run a curated library of AI models already optimized to run on the board's neural processing unit, covering large language models (LLMs), vision-language models (VLMs), speech recognition, and object and gesture detection, right out of the box. They can also bring their own models in the GGUF format from Hugging Face, a popular AI model repository, or train and optimize custom models using the integrated Edge Impulse Studio. LLMs and VLMs can also be deployed with just a few lines of code using Qualcomm's Gen AI Developer Tool, known as GenieX, an open-source runtime for running generative AI directly on-device. The platform has been designed and validated for commercial use across robotics, industrial edge AI, smart cities, AI vision systems, and education, supporting applications ranging from autonomous mobile robots and predictive maintenance to real-time quality inspection and hands-on STEM learning.
Once a design built on VENTUNO Q is ready to move toward commercial production, the same Dragonwing IQ8 processor will soon be available in production-grade System-on-Module (SOM) form from hardware partners, letting developers carry over the software, AI models, and application logic they built during prototyping. SECO, a provider of edge computing and industrial AI solutions, and Toradex, a specialist in industrial-grade SOMs, are among the first partners supporting this transition under the new "Works with Arduino" program.
Nakul Duggal, executive vice president and group general manager for automotive, industrial, and embedded IoT at Qualcomm Technologies, said the Dragonwing IQ8 Series is designed to support the full journey from early development through commercial deployment, and that the "Works with Arduino" program gives developers and partners a practical path to bring products built on Dragonwing IQ8 into actual commercial products, helping speed up development and commercialization across the ecosystem.
Massimo Mauri, CEO of SECO, said the company has spent decades helping businesses turn ideas into products that actually ship and perform in the real world, and that joining the "Works with Arduino" program combines Arduino's accessibility with SECO's production-grade engineering, letting developers prototype quickly and then scale with confidence. Samuel Imgrueth, CEO of Toradex, said VENTUNO Q makes high-performance AI and robotics accessible to a broad developer community, and that Toradex's Aquila architecture, paired with the new Dragonwing IQ8 Series, gives embedded developers a path from prototype to reliable, long-lived industrial and edge AI products.
Beyond prototyping, VENTUNO Q has been designed for commercial deployment across robotics, industrial edge AI, smart cities and connected homes, AI vision systems, and education, supporting applications ranging from autonomous mobile robots and predictive maintenance to real-time quality inspection and hands-on STEM learning.
VENTUNO Q is available to pre-order now through the Arduino Store and official distribution partners including DigiKey, Farnell, Robu.in, Mouser, and RS. Attendees of IFA 2026 will be able to see VENTUNO Q on display at the Qualcomm booth, Hall 22, H22-116.
