Join us as we recap Ultralytics' experience at IOT Solutions World Congress 2025, where we got a chance to showcase how computer vision can power smarter IoT solutions.
Last week, Ultralytics participated in the IoT Solutions World Congress (IOTSWC) 2025 in Barcelona, held from May 13 to 15. The event brought together more than 11,000 visitors from over 100 countries to explore the latest in the Internet of Things (IoT), artificial intelligence (AI), and digital transformation.
IoT refers to the growing network of connected devices that collect and exchange data to improve efficiency, automation, and decision-making. It supports everything from smart homes and cities to manufacturing and healthcare.
Our Machine Learning Engineer, Francesco Mattioli, Hannah Streif, our Business Development Representative, and Zinnia Louise Pourdad, our Events Manager, were on site representing Ultralytics and shared how our computer vision models, like Ultralytics YOLO11, are paving the way to create smarter, more efficient IoT solutions. Simply put, models like YOLO11 can be used to give IoT devices the ability to see so they can detect objects, monitor activity, and respond in real-time.
From keynotes to demos, Ultralytics’ participation at IoT Solutions World Congress was the perfect stage to drive innovation from our shared booth as we partnered with STMicroelectronics, a global leader in semiconductors and IoT technology, to demo YOLO11 running on STM32N6 microcontrollers (MCUs). The live demos showed how Vision AI applications can run efficiently at the edge on low-power devices.
In this article, we’ll share some key moments from our time at IOTSWC 2025, including what we learned, who we connected with, and how Ultralytics is helping drive the future of intelligent and connected technology. Let’s get started!
This year marked the 9th edition of the IoT Solutions World Congress since it first began in 2015. The theme for 2025, “Connect to the Next Level,” focused on how technologies like the IoT, AI, edge computing, and digital twins are transforming industries.
The event featured over 100 sessions led by 150 speakers, including experts like Emanuela Girardi, President of the AI, Data & Robotics Forum, and Brad Morrison, CEO of Atlantis Industries. Talks and panels explored everything from improving operations with AI to innovations in space technology and advanced manufacturing.
Attendees could explore different areas like the main exhibition floor, a startup zone, and immersive demos in “the dome.” The congress also ran alongside the Barcelona Cybersecurity Congress, which focused on keeping digital systems safe in a more connected world.
Altogether, IOTSWC 2025 brought together innovators, businesses, and thought leaders to exchange ideas and explore what’s next for IoT and AI-driven solutions.
We were invited by STMicroelectronics to take part in a session at IOTSWC 2025, where we showed how our Ultralytics YOLO11 models can run directly on their new STM32N6 microcontrollers. These are the first STM32 chips with built-in AI acceleration, designed to handle real-time computer vision tasks on small, low-power devices.
What makes the STM32N6 stand out is that it delivers an experience similar to a microprocessor unit (MPU). An MPU is the type of chip you'd normally find in more powerful systems like smartphones or industrial machines.
However, instead of requiring that level of hardware, you can achieve similar performance on a much smaller and more energy-efficient microcontroller. This results in faster, smarter devices that don’t rely on the cloud.
In the session, we demonstrated how YOLO11 can handle tasks such as object detection, pose estimation, and object tracking, all running directly on the device. For example, in a smart factory setting, this could allow a small sensor-equipped camera to detect workers or moving equipment in real-time, helping improve safety without relying on external servers.
Running everything locally helps reduce latency, improve data privacy, and lower power consumption. These advantages are especially crucial in fields like smart cities, healthcare, and manufacturing, where speed, efficiency, and security matter most.
Next, let’s take a closer look at the session co-hosted by Francesco Mattioli from Ultralytics and Nicolas Gaude, AI Solutions Product Marketing Manager at STMicroelectronics. Titled “Smarter IoT: Computer Vision and Power Efficiency at the Edge,” the 30-minute talk was part of the “Digital Transformation Adoption” track and took place on Tuesday, May 13.
Francesco and Nicolas explored how advances in computer vision and edge AI are enabling a new generation of smarter, more autonomous IoT devices. By running computer vision models like YOLO11 directly on microcontrollers such as the STM32N6, devices can process visual data where it’s captured, on the edge, rather than relying on servers. This shift enables real-time decision-making for tasks like object detection, movement tracking, and gesture recognition.
This concept is particularly important in situations where devices operate in remote or resource-constrained environments. Rather than relying on a constant internet connection or cloud-based processing, edge AI makes it possible for these devices to run vision tasks locally.
The session also touched on performance and efficiency trade-offs, showing how advances in hardware and software optimization are making it possible to run sophisticated AI models on small, energy-efficient chips. Overall, it was a practical look at what’s possible today with edge AI and how it is already being used to build smarter, more responsive systems.
Another key highlight from our time at IOTSWC 2025 was the live demo of Ultralytics YOLO11 at the STMicroelectronics booth, which was one of the largest and most visited booths at the event. Visitors got a close-up look at how real-time computer vision solutions enabled by YOLO11 can run directly on the STM32N6 microcontroller, without relying on cloud processing.
The demo featured tasks like object detection, where the system identifies and labels items such as people, vehicles, or equipment in a video feed, and pose estimation, which tracks key points on the human body to understand movement or posture. Both ran smoothly on a compact, low-power device, demonstrating just how effective and efficient edge AI can be in real-world settings.
Many of the attendees who visited our booth were curious about how real-time object detection could be applied in their specific industries. We spoke with professionals from manufacturing, automotive, retail, shipping, and even sports technology, each exploring how edge-based computer vision could help solve real-world challenges.
Use cases ranged from monitoring machinery on factory floors to tracking athletes' movements to improving inventory visibility and customer experience in retail settings. These conversations made it clear that the ability to run fast, accurate vision AI on small, low-power devices is becoming increasingly relevant across a wide spectrum of applications.
Hearing firsthand how teams are thinking about integrating these capabilities was a great reminder of how flexible and impactful models like Ultralytics YOLO11 can be, especially when paired with efficient hardware like the STM32N6.
Beyond the sessions and demos, IOTSWC 2025 was a great opportunity for us to connect with others in the Vision AI and IoT space - from developers and engineers to product teams and industry partners. We had some great conversations about edge deployment, model optimization, and the practical challenges teams are navigating as they bring AI to devices in the field.
We also caught up with partners and collaborators, including a visit to the Seeed Studio booth, where we exchanged ideas around edge hardware and AI integration. Events like this are always meaningful for strengthening relationships and finding new opportunities for collaboration.
Some of the most memorable moments came from time spent outside the official agenda. On the evening of May 13, the Ultralytics team on the ground enjoyed a team dinner in Barcelona. The following night, the team joined the official IOTSWC after party, where relaxed networking and interactive cocktails wrapped up the day on a lighter note.
The team left the event feeling energized, inspired by the people they met, and excited about the growing momentum behind Vision AI at the edge.
IoT Solutions World Congress 2025 in Barcelona was a great chance for us to connect with other innovators, share our work with STMicroelectronics, and showcase how computer vision is helping make IoT smarter and more efficient.
From live Ultralytics YOLO11 demos to conversations with people across different industries, the event really showed how powerful Vision AI at the edge can be. There was a lot of interest in real-time object detection and edge AI, and it's exciting to see how fast the space is growing. We're looking forward to continuing to build tools that make computer vision easier to use and more impactful for teams everywhere.
Become a part of our growing community! Dive into our GitHub repository to learn more about AI. If you're looking to build computer vision solutions, check out our licensing options. Explore the benefits of computer vision in healthcare and see how AI in logistics is making a difference!
Begin your journey with the future of machine learning
EN
ZH
KO
JA
RU
DE
FR
ES
PT
IT
AR
TR
VI