The Gemini Live Agent Challenge is officially in the books! We challenged developers worldwide to break out of the traditional ‘text box’ paradigm by building next-generation AI agents. From our initial announcement to amassing 11,878 participants and 1,536 submitted projects from 151 countries, the results were nothing short of spectacular.
The mission was to seamlessly integrate multimodal capabilities—building agents that help you see, hear, speak, and create in real time — using the Gemini Live API, the Agent Development Kit (ADK), and the robust infrastructure of Google Cloud. Participants pushed the boundaries of interactive AI across three distinct categories: The Live Agent, The Creative Storyteller, and The UI Navigator.
Congratulations to the builders who took home the top prizes! These winning teams combined technical precision with bold imagination, completely redefining how users can interact with and experience agents. Two of these standout developers were even recognized in person at Google Cloud Next 2026. Here’s a look at their experience, alongside the complete list of winning agents.
Celebrating our category winners at Google Cloud Next ‘26
Category winners Jeremiah Somoine and Bryen Param were invited to attend Google Cloud Next 2026 in Las Vegas, where they shared their experiences and insights with the broader developer community. Both winners presented Lightning Talks at the Developer Theatre on the expo floor and sat down for exclusive interviews in the Creator Studio Pod at the GDE and Certified Lounge.
During his time at the event, Bryen discussed the core inspiration behind drone-copilot. He explained that his project was driven by the question of “what if a model could interact with the real world?”, showcasing how multimodal capabilities can bridge the gap between AI and physical environments.
