How to Beginner to Intermediate · 3 min read

How to use Pydantic AI with OpenAI

Quick answer

Use pydantic-ai by creating a Pydantic model for your expected response and then call client.chat.completions.create from the openai SDK with response_model set to that model. This enables typed, validated AI outputs from OpenAI models like gpt-4o-mini.

PREREQUISITES

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

Setup

Install the required packages and set your OpenAI API key as an environment variable.

Install packages: openai, pydantic-ai, and pydantic.
Set OPENAI_API_KEY in your environment.

bash

pip install openai>=1.0 pydantic-ai pydantic

output

Collecting openai
Collecting pydantic-ai
Collecting pydantic
Successfully installed openai pydantic-ai pydantic-2.x.x

Step by step

Define a Pydantic model for the AI response, initialize the OpenAI client, and call the chat completion with response_model. The AI output is parsed and validated automatically.

python

import os
from openai import OpenAI
import instructor
from pydantic import BaseModel

# Define your Pydantic model for structured output
class User(BaseModel):
    name: str
    age: int

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

# Wrap OpenAI client with pydantic-ai
pydantic_client = instructor.from_openai(client)

# Prepare messages
messages = [{"role": "user", "content": "Extract: John is 30 years old"}]

# Call chat completion with response_model
user = pydantic_client.chat.completions.create(
    model="gpt-4o-mini",
    response_model=User,
    messages=messages
)

print(f"Name: {user.name}, Age: {user.age}")

output

Name: John, Age: 30

Common variations

You can use async calls with await if your environment supports it. Also, you can switch to Anthropic models by using instructor.from_anthropic with the anthropic SDK client. Change the model parameter to use different OpenAI models like gpt-4o or gpt-4o-mini.

python

import asyncio

async def async_example():
    user = await pydantic_client.chat.completions.acreate(
        model="gpt-4o-mini",
        response_model=User,
        messages=[{"role": "user", "content": "Extract: Alice is 25 years old"}]
    )
    print(f"Async Name: {user.name}, Age: {user.age}")

asyncio.run(async_example())

output

Async Name: Alice, Age: 25

Troubleshooting

If you get validation errors, ensure your Pydantic model matches the expected AI output format.
If the API key is missing or invalid, set OPENAI_API_KEY correctly in your environment.
Use the latest openai and pydantic-ai packages to avoid compatibility issues.

✅

Key Takeaways

Use pydantic-ai to parse and validate AI responses with Pydantic models.
Wrap the OpenAI client with instructor.from_openai() for seamless integration.
Always set response_model in chat.completions.create to get typed outputs.
Supports both synchronous and asynchronous usage patterns.
Keep your Pydantic model aligned with the expected AI output structure to avoid validation errors.

Verified 2026-04 · gpt-4o-mini

Verify ↗