Function Calling

Simple Definition

Function calling is a feature of AI APIs that lets a language model output a structured request to call a specific function in your code — instead of just generating text.

When you ask an AI “What’s the weather in Paris right now?”, it can’t know that from its training data. With function calling, it outputs a structured call like get_weather(city="Paris"), your code runs that function, gets the real data, and passes it back to the model to generate a natural response.

How It Works

1. You define functions the model can call, with a description and parameter schema
2. User sends a message that might require one of those functions
3. Model decides whether to generate text or call a function
4. If calling a function, it outputs the function name and parameters in a structured format
5. Your code runs the actual function and returns the result
6. Model uses the result to generate a natural-language response

Why Function Calling Is Important

It’s what bridges the gap between AI text generation and real-world action. Without it, AI can only tell you things. With function calling, AI can:

• Look up current information
• Write to databases
• Send emails or messages
• Create calendar events
• Interact with any external service

Example Use Cases

• Customer service bot that can actually look up your order status
• AI assistant that schedules meetings by accessing your calendar
• Data analysis tool that queries a live database
• AI that can take actions in your app on a user’s behalf

Related Terms

• Tool Use — the broader concept; function calling is the implementation
• AI Agent — agents use function calling to take real-world actions
• API — the external interfaces AI calls via function calling
• LLM — the models that support function calling