Defined term
Function calling
Specific implementation of tool use where the model emits structured JSON calls to registered functions.
Function calling is a model capability where the developer registers a set of functions with JSON schemas, and the model can choose to call them by emitting a structured request. The framework executes the function and returns the result to the model, which continues generation. It is the foundation of tool use in production LLM apps.
When it matters
When you need the model to return structured data or invoke external APIs reliably. Replaces brittle regex parsing of free-text responses with schema-validated calls.
Real example
A meeting-scheduler that exposes 3 functions to Claude: find_available_slots(participants, duration), create_meeting(slot, attendees, agenda), notify_attendees(meeting_id). Model returns structured JSON; framework validates schema and executes.
KPIs to watch
Function-call schema validation rate (>99%), argument accuracy (right values, >95%), retries due to malformed calls (<1%).
Related terms
Tool use
An LLM's ability to call external functions, APIs, or services within a generation step.
Structured output
Constraining model output to a strict schema (JSON, regex, grammar) for reliable downstream parsing.
Agentic AI
AI systems that can plan, take multi-step actions, and use tools to complete tasks autonomously.
Autonomous agent
An AI agent that completes a defined task without per-step human input.
See it in action
We use this every week
Book a 30-min call and we'll walk you through how Function calling shows up in a real engagement we're running.
Book a 30-min call