Google's Interactions API is now GA as the unified way to build with Gemini models and agents, adding managed agents, background execution, and multimodal tools.
Google has announced that the Interactions API has reached general availability and is now the primary interface for building with Gemini models and agents. Launched in public beta in December 2025, the API offers a single unified endpoint with server-side state, background execution, tool combination, and multimodal generation.
Key GA additions include Managed Agents (remote Linux sandboxes with the Antigravity agent), background execution via background=True, mixed built-in and custom tools, Deep Research upgrades, media generation (images, music, speech), a simplified step-based schema, and Flex/Priority pricing tiers. Past interactions are retrievable with 55-day retention on the paid tier.
Google now defaults all AI Studio and Gemini API documentation to the Interactions API. The legacy generateContent API remains supported, but frontier agent capabilities will increasingly land on Interactions first. SDKs are available in Python and JavaScript, with migration guides and a gemini-interactions-api skill for coding agents.
Read more on the Google Blog and the Interactions API documentation.
As the Interactions API reaches general availability, Google positions it as the default front door for Gemini development. Teams building agents no longer need separate integrations for chat history, background jobs, and tool calls — the GA release bundles those paths into one contract. For developers tracking Google's stack, this consolidation mirrors how Antigravity and the Interactions API were pitched together at launch: one interface, many execution modes.
Original Sources
Disclaimer
This is an AI-summarized article from authoritative sources. If you want your content removed, please .