Key highlights from Ultralytics at Embedded Vision Summit 2026

Last week, from May 11 to 12, the Ultralytics team headed to the Santa Clara Convention Center in California, for the Embedded Vision Summit 2026, the premier conference and expo dedicated to practical, deployable computer vision and physical AI. Organized by the Edge AI and Vision Alliance, the Summit has been bringing together the embedded vision community since 2012 and is one of the best places to see where edge AI is heading next.

With around 1,500 attendees from across the ecosystem, including engineers, product creators, and hardware and software teams, the event is uniquely focused on bridging the gap between research and real, shipping products. From live demos and technical sessions to networking and one-on-one conversations, it was a great setting to connect, learn, and share what we've been building.

Our Solutions Engineer, Alexis Schnitger, and Abi Anderson, our Customer Success Manager, were on the ground for two packed days. They demoed Ultralytics YOLO26, the latest model in the Ultralytics YOLO family, sparked conversations with developers and innovators building real-world vision AI products, and connected with key customers and partners.

So let’s walk through some of the highlights from our time at Embedded Vision Summit 2026.

An overview of Embedded Vision Summit 2026

Before we dive into our experience, let's take a closer look at what makes the Embedded Vision Summit such an impactful event for the computer vision and edge AI community.

What sets Embedded Vision Summit apart is its emphasis on practical, deployable solutions. Unlike research-focused conferences, the event is designed for engineers and product teams who are actively building and shipping computer vision products. With over 60 exhibitors showcasing AI processors, accelerators, software tools, and deployment solutions, attendees get a hands-on look at the technologies powering edge AI today.

Beyond the expo floor, Embedded Vision Summit features a rich program of sessions, panels, and demos covering everything from embedded AI and hardware acceleration to real-world case studies and production deployment. It's a space where the conversation goes beyond what's possible and focuses on what's actually being built and shipped.

Inside Embedded Vision Summit 2026

The 2026 edition of Embedded Vision Summit featured a packed program of 90 sessions across four conference tracks, including Fundamentals, Technical Insights, Business Insights, and Enabling Technologies, alongside keynotes, hands-on trainings, workshops, technology exhibits, and networking opportunities.

Keynotes were delivered by Dr. Eric Xing, President of Mohamed bin Zayed University of Artificial Intelligence and a Professor at Carnegie Mellon University, and Dr. Vikas Chandra, Senior Director at Meta Reality Labs. 

Dr. Xing explored how to move AI beyond book intelligence toward systems that can genuinely reason, plan, and act in the real world. Meanwhile, Dr. Chandra made a compelling case that the future of AI isn't about building bigger models, but smarter, more efficient ones that can run directly on-device, bringing real-time understanding to cameras, glasses, and other constrained hardware.

In addition to these sessions, exhibitors showcased processors, algorithms, software, sensors, camera technologies, and development tools for practical, deployable computer vision and edge AI applications. Embedded Vision Summit 2026 also featured the Edge AI and Vision Product of the Year Awards and the AI Innovation Awards, recognizing companies and products making a real impact in physical AI and computer vision.

Key themes at Embedded Vision Summit 2026

Here's a glimpse of some of the key topics that shaped the conversation at this year's Embedded Vision Summit:

• Physical AI: One of the event's biggest themes was physical AI, with sessions exploring how AI systems are moving beyond screens and into the real world. AI enthusiasts are clearly thinking about efficient, edge-native systems that can operate in real environments with real constraints.
• Real-world autonomy: Autonomous systems were a major talking point, with discussions covering humanoid robots, trustworthy self-driving systems, and vision-language-action models for embedded applications. The conversations reflected a growing push to move autonomy out of the lab and into production.
• Prototype to production: A recurring theme throughout Embedded Vision Summit 2026 was the challenge of taking vision AI from a working prototype to a reliable, scalable product. Sessions tackled the real barriers teams face when moving from early pilots to deployment at scale, reflecting how much the community is focused on real-world outcomes.
• Emerging edge AI: From small language models to world models for understanding and predicting environments, Embedded Vision Summit 2026 also looked ahead at what's coming next in edge AI, with a clear direction toward reasoning, planning, and acting beyond simple image recognition.

Before the summit: Arriving in Santa Clara

Before Embedded Vision Summit 2026 officially kicked off, Alexis and Abi flew into San Francisco to settle in ahead of a busy few weeks of meetings. From there, they headed to the Santa Clara Convention Center to get everything ready for the days ahead.

Booth 905 was set up with live demos of Ultralytics YOLO26, giving visitors a hands-on look at a range of computer vision tasks in action. With everything in place, the team was ready to hit the ground running when the doors opened on May 11.

Day one: Demoing YOLO26 and connecting with the community

Day one of Embedded Vision Summit 2026 was all about meeting the community. Alexis and Abi spent the day at our booth, connecting with engineers, product teams, and developers actively building and deploying computer vision solutions.

The live demos of YOLO26 attracted a steady stream of visitors, with object detection and pose estimation drawing particular interest. Attendees had the chance to see these tasks running in real time and explore how they can be applied across a range of real-world use cases in industries such as robotics and manufacturing. 

Many visitors were already using Ultralytics YOLO models in their own projects, which led to insightful conversations about how vision AI is being applied on the ground across various AI solutions.

There was also growing interest in Ultralytics Platform, our newly launched end-to-end solution for computer vision workflows. From data management and annotation to model training, deployment, and monitoring, attendees were keen to learn more about how the platform brings every stage of the computer vision workflow into a single, connected workspace.

Throughout the day, the booth was a great spot for networking, with conversations flowing naturally between demos and discussions.

Day two: From the summit floor to customer conversations

Day two gave our team a chance to explore the expo floor and connect with various partners and customers, picking up where the conversations from day one left off. It was a great opportunity to see what others in the embedded vision space are building and exchange ideas with teams from across the ecosystem.

We also got to connect with STMicroelectronics, one of our key collaborators in the embedded vision space. The meeting was the perfect opportunity to catch up and explore how Ultralytics YOLO models are being used in embedded AI applications on STMicroelectronics hardware.

Overall, it was a fantastic two days of conversations and connections. From there, Alexis and Abi continued their US trip, heading out to meet with Ultralytics customers and partners. 

These in-person visits let us hear directly from the teams building with YOLO models, understand the challenges they are tackling, and see firsthand how vision AI is being put to work in the real world. You can find out more about how our customers are using Ultralytics YOLO models on our customer success stories page.

Key takeaways

Embedded Vision Summit 2026 was a great reminder of how quickly the embedded vision and edge AI community is growing. From live demos and booth conversations to catching up with collaborators and customers, the two days were packed with meaningful interactions and fresh perspectives on where computer vision and AI are heading next.

A big thank you to everyone who stopped by Booth 905, explored a demo, or took the time to chat with us throughout the event. Your curiosity and feedback go a long way in shaping how we continue to build and improve our tools for the vision AI community.

Join our community and GitHub repository to learn more about AI and computer vision. Visit our solutions pages to explore applications like computer vision in agriculture and AI in robotics. Check out our licensing options and get started with computer vision today!