Together Structured Output

Request JSON through Together and validate a MovieScript with Pydantic.

Together retired the source's meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo serverless model. Select a current chat model marked for structured outputs in the Together serverless catalog before running. See Together structured outputs.

This JSON-mode example sends response_format.type="json_object" and validates the response locally. JSON Object mode does not enforce the Pydantic schema on the server.

structured_output.py
"""
Together Structured Output
==========================

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

from typing import List

from agno.agent import Agent, RunOutput  # noqa
from agno.models.together import Together
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!"
    )


# Agent that uses JSON mode
json_mode_agent = Agent(
    model=Together(id="meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo"),
    description="You write movie scripts.",
    output_schema=MovieScript,
    use_json_mode=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 environment variables

export TOGETHER_API_KEY="your_together_api_key_here"
export TOGETHER_STRUCTURED_MODEL_ID="your_current_together_structured_model_id_here"

Use a structured-output model

Add import os, then replace Together(id="meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo") with Together(id=os.environ["TOGETHER_STRUCTURED_MODEL_ID"]) in the saved file.

Run the example

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

python structured_output.py

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