Function Calling
A model feature that allows it to request the execution of developer-defined functions with structured arguments.
Full Definition
Function calling (also called tool calling) lets developers define a set of functions — with names, descriptions, and parameter schemas — and pass them to the model. When the model determines that a function is needed to answer the user, it outputs a structured JSON call with the function name and arguments rather than a prose answer. The developer executes the function and returns the result to the model, which then uses it to form a final response. This enables LLMs to take actions — searching the web, reading files, querying databases, calling APIs — making them practical for agentic applications. OpenAI popularised the feature; Anthropic calls it 'tool use'.
Examples
Defining a get_weather(city: str, unit: str) function; the model calls it with {'city': 'London', 'unit': 'celsius'} when asked about London's weather.
An AI assistant using function calling to fetch a user's calendar events, then summarising the week's schedule in natural language.
Apply this in your prompts
PromptItIn automatically uses techniques like Function Calling to build better prompts for you.
You might also like