Perplexity Structured Output

Request a MovieScript JSON schema from Perplexity sonar-pro and validate the response.

structured_output.py
"""
Perplexity Structured Output
============================

Cookbook example for `perplexity/structured_output.py`.
"""

from typing import List

from agno.agent import Agent, RunOutput  # noqa
from agno.models.perplexity import Perplexity
from pydantic import BaseModel, Field

# ---------------------------------------------------------------------------
# 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!"
    )


# Agent that uses JSON mode
json_mode_agent = Agent(
    model=Perplexity(id="sonar-pro"),
    description="You write movie scripts.",
    output_schema=MovieScript,
    markdown=True,
)

# 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)

json_mode_agent.print_response("New York")

# ---------------------------------------------------------------------------
# Run Agent
# ---------------------------------------------------------------------------

if __name__ == "__main__":
    pass

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 Perplexity API key

export PERPLEXITY_API_KEY="your_perplexity_api_key_here"

Run the example

Save the code above as structured_output.py, then run:

python structured_output.py

Full source: cookbook/90_models/perplexity/structured_output.py