Responses: reasoning.generate_summary interpreted as unsupported parameter

[farfromavocaido]

Confirm this is an issue with the Python library and not an underlying OpenAI API

• [x] This is an issue with the Python library

Describe the bug

When I pass in the reasoning.generate_summary argument (as 'detailed' or 'concise') I get a 400 error:

raise self._make_status_error_from_response(err.response) from None

openai.BadRequestError: Error code: 400 - {'error': {'message': "Unsupported parameter: 'reasoning.generate_summary' is not supported with the 'o3-mini-2025-01-31' model.", 'type': 'invalid_request_error', 'param': 'reasoning.generate_summary', 'code': 'unsupported_parameter'}}

I've tried this with a few (reasoning) models, the same output seems to arise each time.

I can see in the response "reasoning=Reasoning(effort='low', generate_summary=None)" so it's definitely built in somewhere!

To Reproduce

Run a client.responses.create request with any reasoning model and generate_summary set to "detailed" or "concise"

Code snippets

from openai import OpenAI

client = OpenAI()

response = client.responses.create(
  model="o3-mini",
  input="Can you outline how to determine the capital of France.",
  reasoning={
        "effort": "medium",
        "generate_summary": "detailed"
        }
)

print(response)

OS

macOS

Python version

python 3.12.2

Library version

openai v1.66.3