How to beginner · 3 min read

How to use function calling with Claude API

Q: How to use function calling with Claude API

Use the Anthropic Python SDK to call Claude models with the functions parameter specifying your function schema. The API returns structured JSON in message.function_call.arguments for easy parsing and integration.

Quick answer

Use the Anthropic Python SDK to call Claude models with the functions parameter specifying your function schema. The API returns structured JSON in message.function_call.arguments for easy parsing and integration.

PREREQUISITES

Python 3.8+
Anthropic API key
pip install anthropic>=0.20

Setup

Install the anthropic Python SDK and set your API key as an environment variable.

bash

pip install anthropic>=0.20

Step by step

This example demonstrates how to call a Claude model with a function schema to get structured JSON output representing a user profile.

python

import os
from anthropic import Anthropic

client = Anthropic(api_key=os.environ["ANTHROPIC_API_KEY"])

function_schema = {
    "name": "create_user_profile",
    "description": "Create a user profile with name, age, and email.",
    "parameters": {
        "type": "object",
        "properties": {
            "name": {"type": "string", "description": "Full name of the user"},
            "age": {"type": "integer", "description": "Age of the user"},
            "email": {"type": "string", "description": "Email address"}
        },
        "required": ["name", "age", "email"]
    }
}

response = client.messages.create(
    model="claude-3-5-sonnet-20241022",
    system="You are a helpful assistant that outputs structured JSON.",
    messages=[{"role": "user", "content": "Generate a user profile for Alice, age 30, email alice@example.com."}],
    functions=[function_schema],
    function_call={"name": "create_user_profile"},
    max_tokens=300
)

# Extract the structured JSON arguments from the function call
structured_output = response.content[0].function_call.arguments
print(structured_output)

output

{
  "name": "Alice",
  "age": 30,
  "email": "alice@example.com"
}

Common variations

Use different Claude models like claude-sonnet-4-5 for improved coding or reasoning.
Call multiple functions by passing a list of function schemas in functions.
Use function_call="auto" to let Claude decide which function to call based on the prompt.
Implement async calls with asyncio and client.messages.acreate() for concurrency.

Troubleshooting

If you get a 400 Bad Request, verify your function schema is valid JSON Schema.
If function_call.arguments is empty, ensure function_call parameter matches a function name in functions.
Check your API key environment variable ANTHROPIC_API_KEY is set correctly.

✅

Key Takeaways

Use the functions parameter to define JSON Schema for structured outputs with Claude.
Set function_call to specify which function Claude should invoke for structured data.
Parse response.content[0].function_call.arguments to get the structured JSON result.
Use claude-3-5-sonnet-20241022 or claude-sonnet-4-5 for best coding and structured output support.
Validate your function schema and environment variables to avoid common errors.

Verified 2026-04 · claude-3-5-sonnet-20241022, claude-sonnet-4-5

Verify ↗