Structured Output Using Pydantic

In this lesson, you’ll learn how to extract structured data from user input using a Pydantic model with OpenAI’s parse() method.

Prerequisites

Install the required packages:

pip install openai pydantic

The Code

from openai import OpenAI
from pydantic import BaseModel

client = OpenAI()

class ContactInfo(BaseModel):
    name: str
    email: str
    phone: str

response = client.responses.parse(
    model="gpt-4o-mini",
    input=[
        {"role": "system", "content": "Extract contact details from the message."},
        {
            "role": "user",
            "content": "You can reach out to John Doe at john@example.com or call him at +1-555-1234."
        },
    ],
    text_format=ContactInfo,
)

contact = response.output_parsed

print(f"Name: {contact.name}")
print(f"Email: {contact.email}")
print(f"Phone: {contact.phone}")

Explanation

• BaseModel: Defines the expected output format using Pydantic.
• client.responses.parse(...): Parses the model output directly into the ContactInfo object.
• response.output_parsed: Structured data you can use immediately.

Use Case

Best for:

• Information extraction (contacts, product specs, summaries)
• Creating structured datasets from unstructured prompts
• Automating form-filling and data ingestion