Today, the Interactions API has been announced as generally available, serving as the primary interface for interacting with Gemini models and agents. This release, which follows a successful public beta launched in December 2025, incorporates significant enhancements based on developer feedback, including features like Managed Agents and background execution. As the API stabilizes, all documentation will now default to it, and efforts are underway to ensure its adoption as the standard interface in third-party SDKs and libraries.

Gemini: Gemini refers to Google’s family of AI models and agents designed for inference, multimodal tasks, and autonomous workflows. It forms the core technology supported by the Interactions API, enabling developers to build applications through simple calls for models or agents. The GA release emphasizes Gemini’s role in agentic development with upcoming features like Gemini Omni.
Interactions API: The Interactions API is the primary developer interface for interacting with Gemini models and agents. It reached general availability after a public beta launch in December 2025, featuring a stable schema and optimizations for stateful, agentic workflows. The API now serves as the default interface for Gemini, with updated documentation and plans to integrate it across third-party SDKs and libraries.

API Release: The Interactions API has reached general availability and is now the primary interface for Gemini models and agents.
Developer Feedback: The API was built based on developer feedback and optimized for stateful, agentic workflows including Managed Agents and background execution.
Ecosystem Integration: All documentation now defaults to the Interactions API, with ongoing work to make it the default across third-party SDKs and libraries.