Python: [Bug]: Formats are not working.

[TahaPasa]

Description

I am trying to use a structured format with agents, but it seems not to be working. I also checked out the code of the client, but I could not find any kind of "format" or "response_format" arguments.

Code Sample

async def run_my_agent(transcript : str):
    agent = client.as_agent(
        name="InfoExtractor",
        instructions=SYSTEM_PROMPT.agent_prompt
    )
    response = await agent.run(
        f"Please provide structured information about the person based on the talks:  '{transcript}'",
        response_format=PersonInfo,
    )
    
    return response

Error Messages / Stack Traces

None of the overloads for "run" match the provided arguments. Argument types: (str, type[PersonInfo])

Package Versions

agent-framework: 1.0.1

Python Version

Python 3.13

Additional Context

No response

[moonbox3]

I can see that our docs are not correct in how this works - I will fix this now.

Our samples do show the correct way to handle this - one needs to either pass in the options dict during the run, or as default_options dict as part of the agent constructor.

Here's an example of passing the Pydantic model at runtime:

agent-framework/python/samples/02-agents/providers/azure/openai_client_with_structured_output.py

Line 44 in 5777ed2

result = await agent.run(query, options={"response_format": OutputStruct})