Definition
Function calling is a capability of large language models that allows them to generate structured JSON outputs matching predefined function signatures, enabling them to invoke external tools and APIs as part of their reasoning process.
In Depth
Function calling was introduced to bridge the gap between natural language understanding and structured tool invocation. When an LLM supports function calling, developers define functions with names, descriptions, and parameter schemas. The model then determines when to call a function, generates the appropriate arguments as JSON, and the application executes the function and returns the result. This is the foundation of AI agent architectures where LLMs need to interact with external systems. Search APIs like Scavio provide pre-built function definitions for popular frameworks, making it trivial to add web search capabilities to any function-calling-enabled LLM.
Example Usage
A developer defines a 'search_google' function with parameters for query, location, and language. When a user asks the chatbot about current events, the LLM generates a function call with the appropriate search query, the app executes it via Scavio, and the LLM summarizes the results.
Platforms
Function Calling (LLM) is relevant across the following platforms, all accessible through Scavio's unified API:
- Amazon
- YouTube
- Walmart
Related Terms
AI Agent Tool Calling
Tool calling is the mechanism by which an AI agent instructs a large language model to invoke an external function or AP...
Model Context Protocol (MCP)
Model Context Protocol (MCP) is an open standard that defines how large language models discover and invoke external too...
Agent Orchestration Framework
An agent orchestration framework is a software library that manages the lifecycle, communication, and tool usage of one ...