.svg){kind=spacer}
Details
- Google released new Gemini API tooling capabilities that let developers combine custom function calling with built-in tools like Google Search in a single API call, streamlining multi-tool workflows.
- The update introduces hybrid tool usage, allowing developers to leverage both proprietary functions and Google's integrated tools without separate API calls or context switching.
- Cross-tool context circulation enables information to flow between custom and built-in tools, improving reasoning across chained operations.
- Tool response IDs provide better tracking and management of outputs from multiple tools operating in parallel or sequence.
- Available now via the Gemini API through Google AI Studio and Vertex AI for developers building with Gemini 3.1 Pro and other current models.
Impact
This release addresses a key developer friction point in agentic AI workflows. Previously, combining external APIs with Google's search capabilities required orchestrating multiple API calls and manually passing context between systems, increasing latency and error surface area. By enabling hybrid tool usage within a single call, Google reduces architectural complexity and makes it easier for developers to build multi-step agents that reason across diverse data sources. The feature appears positioned to compete with OpenAI's parallel function calling in GPT-4 by offering native integration with search—a capability GPT-4 requires third-party tools to match. This matters as enterprises increasingly demand agents that can autonomously research, validate, and act on information; tighter tool composition lowers the barrier to deployment. Over the next 12 months, expect this pattern to drive adoption of Gemini in production agent systems, particularly in customer service and research automation where search context is critical to decision quality.
