AI Dictionary › Prompting

Function Calling

Function calling is a model's ability to output, instead of free text, a structured call to a function you have defined, with a name and arguments in JSON. It does not execute the function: it emits the intent to call it, and you (or your code) run it and return the result. Example: you define get_weather(city, date); the user asks "what's the weather in Rome tomorrow?" and the model returns {"name":"get_weather","arguments":{"city":"Rome","date":"2025-07-19"}}. Your backend calls the real weather API and feeds the result back for the final answer.

Definition

Use it to connect the model to tools, APIs and databases reliably: bookings, calculations, queries, automations. It is the technical foundation of agents and replaces brittle free-text parsing with a structured interface.

Related terms

• Tool Use
• Structured Output
• JSON Mode
• ReAct (Reasoning + Acting)
• Output Parsing

More in Prompting

• Prompt
• Prompting
• Role Assignment
• Zero-shot
• Few-shot
• Chain of Thought
• Prompt Engineering
• Output Format

Put it into practice

• Real scenario: apply this concept with an instant AI verdict.
• Daily gym: short drills to train dictionary concepts.
• Playoff game: challenge other players, elimination-style.

📱 Download the Android app (beta) iOS coming soon

Say what you mean. Get what you need.

Grace Certified, the AI coach that trains and certifies your prompt engineering, by Agora Intelligence.