Google launches Gemini Interactions API for developers

Google DeepMind announced on June 22 that its Interactions API for Gemini models and agents is now generally available (GA). The API, previously in public beta, becomes the primary interface for text generation, streaming, multi‑turn chat, multimodal inputs (images, audio, video, documents), image and speech generation, structured JSON output, tool use, managed agents and background execution.

Managed agents provide a remote Linux sandbox for tasks such as web browsing and file management, while the background=True setting enables asynchronous server‑side processing and integration with Google Search and Maps. New pricing tiers let developers choose between cost‑efficiency and low latency, and a 55‑day interaction history is now available. The service is accessible through Google AI Studio and SDKs for Python and JavaScript, with migration guides for existing generateContent users.