Output response types are not usable as parameters

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

Consider the following code:

messages: ResponseInputParam = [{"role": "developer", "content": instructions}]

# ...
resp = client.responses.create(
    model="o4-mini",
    input=messages,
)

for item in resp.output:
    match item:
        case ResponseOutputMessage(content=content):
            print(item.content)
            messages.append(item)

This seems like it should work fine, but it generates a type checking error:

error: Argument of type "ResponseOutputMessage" cannot be assigned to parameter "object" of type "ResponseInputItemParam" in function "append"
    Type "ResponseOutputMessage" is not assignable to type "ResponseInputItemParam"
      "ResponseOutputMessage" is not assignable to "EasyInputMessageParam"
      "ResponseOutputMessage" is not assignable to "Message"
      "ResponseOutputMessage" is not assignable to "ResponseOutputMessageParam"
      "ResponseOutputMessage" is not assignable to "ResponseFileSearchToolCallParam"
      "ResponseOutputMessage" is not assignable to "ResponseComputerToolCallParam"
      "ResponseOutputMessage" is not assignable to "ComputerCallOutput"
      "ResponseOutputMessage" is not assignable to "ResponseFunctionWebSearchParam"
    ... (reportArgumentType)

I think this is because the parameter types use Required, but the output types don't. For example, EasyMessageInputParam requires content, but ResponseOutputMessage doesn't:

https://github.com/openai/openai-python/blob/ed53107e10e6c86754866b48f8bd862659134ca8/src/openai/types/responses/easy_input_message_param.py#L13-L18

vs

https://github.com/openai/openai-python/blob/ed53107e10e6c86754866b48f8bd862659134ca8/src/openai/types/responses/response_output_message.py#L20-L21

I recognize that the bindings are auto-generated and this may not be trivial to fix.

To Reproduce

n/a

Code snippets

OS

linux

Python version

Python v3.13.2

Library version

openai v1.75.0

Start by comparing src/openai/types/responses/easy_input_message_param.py and src/openai/types/responses/response_output_message.py, focusing on the required fields described in the report. Trace how these auto-generated types are produced before deciding where the compatibility change belongs. Done means the reported messages.append(item) pattern passes Python type checking without weakening the intended response and input type requirements.