Gemini API: Tool and Function Calling

Leverage the function calling feature in Gemini API to build integrations between AI models and their environments—turning simple chatbots into powerful agentic assistants. Instructor Fikayo Adepoju begins by discussing what function calling is all about: the benefits, use cases, and its relationship and semblance with tool calling. He then dives deep into the art of defining functions for models to suggest and generate arguments for, and then explores different modes that give you more control over how the function calling operation goes. From there, he shows how to define, generate, and invoke functions. Fikayo wraps up by covering how to use grounding with Google Search to ensure accurate responses, as well as best practices for implementing function calling. Tune in and start taking advantage of function calling in your Gemini API applications.

Learning objectives
Describe the use and importance of function calling on the Gemini API.
Demonstrate how to configure function calling in model requests.
Declare a pool of functions that are available for the model to suggest.
Discuss how to work with single-turn and multiturn function calling strategies.
Use tools to search the web.
Dynamically decide when to ground the model responses with web search.