How to beginner · 3 min read

How to use function calling with OpenAI API

Q: How to use function calling with OpenAI API

Use the functions parameter in client.chat.completions.create to define callable functions with JSON schemas. The model returns structured data invoking these functions, enabling precise control over output format and content.

Quick answer

Use the functions parameter in client.chat.completions.create to define callable functions with JSON schemas. The model returns structured data invoking these functions, enabling precise control over output format and content.

PREREQUISITES

Python 3.8+
OpenAI API key (free tier works)
pip install openai>=1.0

Setup

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

bash

pip install openai>=1.0

Step by step

Define your function schema and call the OpenAI chat completion endpoint with the functions parameter. The model will respond with a function_call message containing arguments matching your schema.

python

import os
from openai import OpenAI

client = OpenAI(api_key=os.environ["OPENAI_API_KEY"])

functions = [
    {
        "name": "get_current_weather",
        "description": "Get the current weather in a given location",
        "parameters": {
            "type": "object",
            "properties": {
                "location": {"type": "string", "description": "The city and state, e.g. San Francisco, CA"},
                "unit": {"type": "string", "enum": ["celsius", "fahrenheit"]}
            },
            "required": ["location"]
        }
    }
]

messages = [
    {"role": "user", "content": "What is the weather like in Boston?"}
]

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=messages,
    functions=functions,
    function_call="auto"  # or specify function name
)

message = response.choices[0].message

if message.get("function_call"):
    function_name = message["function_call"]["name"]
    arguments = message["function_call"]["arguments"]
    print(f"Function to call: {function_name}")
    print(f"Arguments: {arguments}")
else:
    print(message.content)

output

Function to call: get_current_weather
Arguments: {"location": "Boston"}

Common variations

Use function_call="none" to disable function calling and get normal chat completions.
Specify function_call="get_current_weather" to force calling a specific function.
Use async calls with asyncio and await if your environment supports it.
Switch models like gpt-4o-mini for faster, cheaper calls.

Troubleshooting

If you get a 400 Bad Request, verify your function JSON schema is valid and matches OpenAI's specification.
If the model does not return a function_call, check that function_call parameter is set to "auto" or the function name.
Ensure your API key is correctly set in os.environ["OPENAI_API_KEY"].

✅

Key Takeaways

Use the functions parameter to define callable functions with JSON schemas for structured outputs.
Set function_call to control when and which function the model should invoke.
Parse the function_call field in the response to extract function name and arguments.
Validate your function schema to avoid request errors.
You can combine function calling with normal chat messages for flexible workflows.

Verified 2026-04 · gpt-4o-mini

Verify ↗