GPT5 Models | Tool choices other than 'auto' are not supported with model

[hayescode]

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

• This is an issue with the Python library

Describe the bug

We are unable to use any tool_choice besides 'auto' for GPT5 models in the Responses API using the latest release v.1.99.5.

Documentation: https://platform.openai.com/docs/guides/latest-model#allowed-tools

BadRequestError: Error code: 400 - {'error': {'message': "Tool choices other than 'auto' are not supported with model 'gpt-5-mini-2025-08-07' and the following tool types: 'code_interpreter', 'file_search'.", 'type': 'invalid_request_error', 'param': 'tool_choice', 'code': None}}

https://platform.openai.com/docs/api-reference/responses/create#responses_create-tool_choice

https://community.openai.com/t/error-400-tool-choices-other-than-auto-are-not-supported-with-model-gpt-5-and-the-following-tool-types-web-search-preview/1337152

https://platform.openai.com/docs/guides/latest-model#allowed-tools

To Reproduce

stream = await llm.responses.create(
    model="gpt-5",
    input="Use your python tool to do this math: 8*9183*7663",
    tools=[
        {"type": "code_interpreter", "container": {"type": "auto"}},
    ],
    tool_choice={
        "type": "allowed_tools",
        "mode": "auto",
        "tools": [
            {
                "type": "code_interpreter",
            }
        ],
    },
    stream=True,
)
async for event in stream:
    print(event.model_dump_json(indent=2))

Code snippets

OS

Windows 11

Python version

Python 3.13.2

Library version

openai v.1.99.5

[hayescode]

Also, with GPT 5 we can't set reasoning summary to 'concise'

openai-python/src/openai/types/shared_params/reasoning.py

Lines 31 to 36 in 7aa3c78

	summary: Optional[Literal["auto", "concise", "detailed"]]
	"""A summary of the reasoning performed by the model.
	This can be useful for debugging and understanding the model's reasoning
	process. One of `auto`, `concise`, or `detailed`.
	"""

stream = await llm.responses.create(
    model="gpt-5-mini",
    input="what is a recursive python function?",
    reasoning={"effort": "medium", "summary": "concise"},
    stream=True
)
async for event in stream:
    print(event.model_dump_json(indent=2))

BadRequestError: Error code: 400 - {'error': {'message': "Unsupported value: 'concise' is not supported with the 'gpt-5-mini-2025-08-07' model. Supported values are: 'detailed'.", 'type': 'invalid_request_error', 'param': 'reasoning.generate_summary', 'code': 'unsupported_value'}}

[hayescode]

@RobertCraigie do you know if this is being addressed in an upcoming release? This is our last blocker for converting our applications to GPT-5.