Google DeepMind Upgrades Gemini API with Multi-Tool Chaining and Context Circulation

Google DeepMind has launched a significant upgrade to the Gemini API, allowing developers to combine multiple tools in a single request — something that previously required separate calls and manual handling between steps.

What's New?

The headline feature is multi-tool chaining: built-in tools like Google Search and Google Maps can now be combined with custom functions in a single API request. Results from one tool are automatically passed to the next through what Google calls context circulation.

Each tool call also receives a unique ID, making debugging significantly easier for developers.

Google Maps as a Data Source

A notable addition is that Google Maps is now available as a direct data source for the Gemini 3 model family. This enables location data, business information, and commute times directly in AI workflows — without additional API integrations.

Interactions API

Google recommends the new Interactions API for building these composite workflows. This is part of Google's push into agentic AI where the model can perform multiple actions sequentially without human intervention.

What This Means for CIOs

For businesses building AI agents, this is a clear improvement: lower latency, simpler code, and more reliable debugging. The Google Maps integration in particular opens up interesting use cases in logistics, real estate, and field service.