Updated · The Keyword | Google Product and Technology News · Jun 22
Google Makes Interactions API Primary Gemini Interface, Cutting Flex Tier Costs 50%
Updated
Updated · The Keyword | Google Product and Technology News · Jun 22
Google Makes Interactions API Primary Gemini Interface, Cutting Flex Tier Costs 50%
2 articles · Updated · The Keyword | Google Product and Technology News · Jun 22
Summary
General availability makes the Interactions API Google’s primary way to access Gemini models and agents, with a stable schema and documentation now defaulting to it across Google AI Studio and the Gemini API.
New capabilities added since the December 2025 beta include Managed Agents, asynchronous background execution, richer tool use, Deep Research upgrades, and media generation spanning images, music and multi-speaker text-to-speech.
Flex and Priority tiers let developers trade off cost and latency, with Flex reducing costs by 50%; paid users also get 55-day interaction retention and more precise field-level error reporting.
The API’s schema shifts from legacy roles to typed steps for stateful agent workflows, while Google says frontier features for long-running models and agents will increasingly land only on Interactions API.
Google said the legacy generateContent API will remain supported and keep receiving mainline Gemini models, but it is steering new projects and partner SDKs toward the newer interface.