Google DeepMind has transitioned the Interactions API to its default interface for Gemini models and agents, marking a significant shift in how developers interact with these systems. The API, which was in beta since December 2025, is now generally available and replaces the older generateContent interface used in Google AI Studio and all documentation. While the previous API remains functional, new agent features will only be available through the Interactions API moving forward. 'Interactions sets the stage for the new era of Agents,' wrote Google's developer relations lead, Logan Kilpatrick. Developers are encouraged to follow the migration guide provided by Google to adapt to the new interface.
The updated API includes several enhancements, such as Managed Agents with their own Linux sandbox, background execution for long-running tasks, tool chaining with Google Search and Maps, and media generation for images, music, and speech. Google also simplified the schema since the beta phase, replacing the old role structure with labels like 'user' and 'model' with a new system of typed steps. Each action, from user input to function calls, is now a defined step. Developers can choose between Flex and Priority mode, with Flex reducing costs by 50 percent and Priority optimizing for speed.
According to the source, Google has published a migration guide for developers to transition to the Interactions API. The company has also emphasized the importance of the new API in enabling advanced agent functionalities. The change is part of Google's broader strategy to improve the efficiency and capabilities of its AI models.
Source: thedecoder
