How to beginner · 3 min read

How to use function calling with Gemini API

Q: How to use function calling with Gemini API

Use the OpenAI SDK v1 to call the Gemini API with function_call parameters in chat.completions.create. Define your functions schema and pass it in the request to receive structured function call responses from the gemini-2.5-pro model.

Quick answer

Use the OpenAI SDK v1 to call the Gemini API with function_call parameters in chat.completions.create. Define your functions schema and pass it in the request to receive structured function call responses from the gemini-2.5-pro model.

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

This example shows how to define a function schema and invoke the Gemini API with function calling enabled to get structured outputs.

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"]
        }
    }
]

response = client.chat.completions.create(
    model="gemini-2.5-pro",
    messages=[{"role": "user", "content": "What is the weather like in New York?"}],
    functions=functions,
    function_call="auto"  # Let the model decide to call the function
)

print("Model response:")
print(response.choices[0].message.content)
print("Function call details:")
print(response.choices[0].message.function_call)

output

Model response:

{"location": "New York", "unit": "fahrenheit"}

Function call details:
{'name': 'get_current_weather', 'arguments': '{"location": "New York", "unit": "fahrenheit"}'}

Common variations

Use function_call="none" to disable function calling.
Specify function_call={"name": "function_name"} to force a specific function call.
Use other Gemini models like gemini-2.0-flash for faster responses.
Implement async calls with asyncio and await in Python 3.8+.

Troubleshooting

If you get a 400 Bad Request, verify your function schema is valid JSON Schema.
If the function call is missing, ensure functions and function_call parameters are correctly passed.
Check your API key and environment variable OPENAI_API_KEY are set properly.

✅

Key Takeaways

Use the OpenAI SDK v1 with functions and function_call parameters to enable Gemini API function calling.
Define your function schema using JSON Schema to get structured, machine-readable outputs.
You can let the model decide which function to call or force a specific function call.
Always validate your function schema and environment setup to avoid request errors.

Verified 2026-04 · gemini-2.5-pro

Verify ↗