Cancel for streaming Responses

Confirm this is a feature request for the Python library and not the underlying OpenAI API.

• This is a feature request for the Python library

Describe the feature or improvement you're requesting

Allow for cancellation of streaming/synchronous responses.

Additional context

Currently, responses.cancel is only supported for background responses

For streaming responses one can sever the connection and start a new run from a new response_id node.

However, this could have cost implications. I.e. if a user cancels a run, we do not want to pay for extra compute on the the next ResponseCompletedEvent; this is especially notable if the LLM wants to do multiple tool calls.

Is this, canceling a streaming response, something that is already happening by default when a an OpenAI user disconnects? if not could we please have this feature?

Start in src/openai/resources/responses/responses.py around the responses.cancel definition at lines 1457–1458, then trace how streaming responses are handled. Determine whether disconnecting a stream already cancels server-side work or whether library support is needed. Done means a streaming or synchronous response can be cancelled without continuing unnecessary processing, with behavior confirmed by tests or documented API behavior.