Support for truncation_strategy, max_prompt_tokens and max_completion_tokens

[passivebook]

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

Can the python package be updated to support truncation_strategy, max_prompt_tokens and max_completion_tokens as seen in the OpenAI Documentation here?

Is it possible to do this already?

Additional context

No response

[CoderTonyB]

I'm having the same problem. I need to upload a few thousand words as part of my workflow and it's truncating them. I see truncation_strategy as a header, but I can't figure out where to add them. As shown here:
https://platform.openai.com/docs/api-reference/runs/createRun#runs-createrun-truncation_strategy

I've tried using extra_headers:

assistant = client.beta.assistants.create(
    name="PDF Extract-O-Matic",
    instructions=instruction,
    tools=[{"type": "code_interpreter"}],
    model="gpt-4-turbo-2024-04-09",
    extra_headers={"truncation_strategy": "last_messages", "last_messages": "10"},
)

but when I add a message it still truncates the middle.

My workaround right now is just to upload the larger text as a file and reference it. It would be ideal if I didn't have to do that.

[rattrayalex]

I believe truncation_strategy is a body parameter, not a header.

This will be added to the SDKs shortly; thank you for your patience.

[pstern-sl]

This is available in version 1.18 released today #1315