Google DeepMind's Interactions API Reaches General Availability for Gemini Models

On July 14, 2026, Google DeepMind announced the Interactions API reached general availability, establishing it as the new primary interface for all Gemini models and agents and replacing the legacy generateContent API. For broader context, explore our AI Tools Pricing.

Interactions API: A Unified Approach to AI Development

The Interactions API, which first launched in public beta in December 2025, is designed to simplify the process of building sophisticated AI applications. By consolidating various functionalities, Google DeepMind seeks to lower the technical barrier for developers working with Gemini models and agents. This unified approach is intended to foster more efficient development workflows and enable the creation of more complex AI systems.

Key Features of the General Availability Release

The general availability release introduces several notable enhancements:

• Managed Agents: These agents now include remote Linux sandboxes, which can be provisioned with a single API call. This feature supports reasoning, code execution, web browsing, and file management within a secure environment.
• Background Execution: Interactions can now run asynchronously, with data retrieval retained for 55 days on the paid tier. This mode is suitable for long-running tasks that do not require immediate responses.
• Tool Combination: The API facilitates the integration and combination of various tools, enhancing the agents' capabilities.
• Upgraded Deep Research: This feature now offers two distinct agent versions, providing more flexibility for research-oriented applications.
• Multimodal Media Generation: The API supports the generation of various media types, including images via Nano Banana 2, music through Lyria 3, and expressive multi-speaker text-to-speech (TTS).

Cost Reduction and Accessibility

Google DeepMind has also introduced a new Flex tier, which offers a 50% cost reduction compared to standard pricing. This aims to make the Interactions API more accessible to a broader range of developers and projects. Furthermore, all Google AI Studio and Gemini API documentation now defaults to the Interactions API, indicating its central role in Google's AI ecosystem.

Implications for Developers

The shift to the Interactions API as the primary interface means developers currently using the generateContent API will need to transition to the new system. This change is positioned as an upgrade, offering more integrated and powerful tools for building AI applications. The inclusion of features like Managed Agents with remote sandboxes and background execution provides new avenues for creating more autonomous and complex AI behaviors.

Conclusion

The general availability of Google DeepMind's Interactions API marks a significant step in the evolution of the Gemini ecosystem. By unifying key functionalities and introducing features like Managed Agents and multimodal generation, Google aims to empower developers to build more advanced and efficient AI applications. The introduction of a Flex tier also suggests a commitment to broader accessibility, potentially accelerating the adoption of Gemini models across various development projects.

Sources

• GitHub - google-deepmind/lab: A customisable 3D platform for agent-based AI research · GitHub
• Gemini Robotics 1.5 brings AI agents into the physical world — Google DeepMind
• Start building with Nano Banana 2 Lite and Gemini Omni Flash
• https://blog.google/innovation-and-ai/technology/developers-tools/interactions-api-general-availability/