Chat Structured Output

The source's rating: Dict[str, int] emits an open-ended object and is omitted from the generated required fields. That violates OpenAI's strict schema contract. Apply the typed ratings fix below before running the strict examples.

structured_output.py

"""
Openai Structured Output
========================

Cookbook example for `openai/chat/structured_output.py`.
"""

import asyncio
from typing import Dict, List

from agno.agent import Agent, RunOutput  # noqa
from agno.models.openai import OpenAIChat
from pydantic import BaseModel, Field
from rich.pretty import pprint  # noqa

# ---------------------------------------------------------------------------
# Create Agent
# ---------------------------------------------------------------------------


class MovieScript(BaseModel):
    setting: str = Field(
        ..., description="Provide a nice setting for a blockbuster movie."
    )
    ending: str = Field(
        ...,
        description="Ending of the movie. If not available, provide a happy ending.",
    )
    genre: str = Field(
        ...,
        description="Genre of the movie. If not available, select action, thriller or romantic comedy.",
    )
    name: str = Field(..., description="Give a name to this movie")
    characters: List[str] = Field(..., description="Name of characters for this movie.")
    storyline: str = Field(
        ..., description="3 sentence storyline for the movie. Make it exciting!"
    )
    rating: Dict[str, int] = Field(
        ...,
        description="Your own rating of the movie. 1-10. Return a dictionary with the keys 'story' and 'acting'.",
    )


# Agent that uses JSON mode
json_mode_agent = Agent(
    model=OpenAIChat(id="gpt-5.6-luna"),
    description="You write movie scripts.",
    output_schema=MovieScript,
    use_json_mode=True,
)

# Agent that uses structured outputs with strict_output=True (default)
structured_output_agent = Agent(
    model=OpenAIChat(id="gpt-5.6-luna"),
    description="You write movie scripts.",
    output_schema=MovieScript,
)

# Agent with strict_output=False (guided mode)
guided_output_agent = Agent(
    model=OpenAIChat(id="gpt-5.6-luna", strict_output=False),
    description="You write movie scripts.",
    output_schema=MovieScript,
)

# Get the response in a variable
# json_mode_response: RunOutput = json_mode_agent.run("New York")
# pprint(json_mode_response.content)
# structured_output_response: RunOutput = structured_output_agent.run("New York")
# pprint(structured_output_response.content)

# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------
if __name__ == "__main__":
    # --- Sync ---
    json_mode_agent.print_response("New York")

    structured_output_agent.print_response("New York")

    guided_output_agent.print_response("New York")

    # --- Sync + Streaming ---
    structured_output_agent.print_response("New York", stream=True)

    # --- Async + Streaming ---
    async def main():
        await structured_output_agent.aprint_response("New York", stream=True)

    asyncio.run(main())

Required Schema Fix

In your saved copy, define Rating before MovieScript:

rating_model.py

class Rating(BaseModel):
    story: int = Field(..., description="Story rating from 1 to 10.")
    acting: int = Field(..., description="Acting rating from 1 to 10.")

Replace the entire rating: Dict[str, int] = Field(...) declaration inside MovieScript with:

movie_rating_field.py

rating: Rating = Field(..., description="Your own ratings of the movie.")

Keep the existing imports, other fields, agents and run calls. JSON mode requests valid JSON; native structured output sends JSON Schema, with strict enforcement enabled by default. strict_output=False relaxes provider enforcement. The descriptions ask for ratings from 1 to 10 but do not enforce that range locally.

output_schema describes the expected type. If parsing or validation fails, result.content can remain a string. Before accessing schema fields in a run result, use isinstance(result.content, YourSchema), replacing YourSchema with the class you passed as output_schema.

Run the Example

Set up your virtual environment

uv venv --python 3.12
source .venv/bin/activate

Install dependencies

uv pip install -U agno openai

Export your OpenAI API key

export OPENAI_API_KEY="your_openai_api_key_here"

Run the example

Save the source as structured_output.py, apply the required schema fix above, then run:

python structured_output.py

Full source: cookbook/90_models/openai/chat/structured_output.py