Create Message and create Run Assistant take too long time

API
1

Describe the bug

From the API I had constructed a number of calls to the Wizards I had created. However, as of today, when I run the following command:
client.beta.threads.messages.create(thread_id=thread.id, role="user", content=user_message)
and
client.beta.threads.runs.create(thread_id=thread.id,assistant_id=assistant_id)
Upon examination of the library during debugging, a significant loop was identified. However, due to my limited understanding of the library, I am unable to discern its nature.
The prolonged time required to create a message and a Run, coupled with the status of ‘expired’ assigned to the Run upon reaching the execution point of a calling function, has resulted in the unavailability of the implementation, which is already operational in a production environment.

Thanks a lot for the hard work maintaining this code!

To Reproduce

#Create any thread
thread = client.beta.threads.create()
client.beta.threads.messages.create(thread_id=thread.id, role="user", content=user_message) # Here the code takes more than 1 minute
client.beta.threads.runs.create(thread_id=thread.id,assistant_id=assistant_id,) # Here the code take more than 9 minutes

Code snippets

def get_response(client, thread):
    return client.beta.threads.messages.list(thread_id=thread.id, order="asc")


def process(client, assistant_id, question):
    time1 = time.time()
    print("init")
    thread, run = create_thread_and_run(client, assistant_id, question)
    print(f"Thread and run created in {time.time() - time1} seconds")
    run = wait_on_run(client, run, thread)
    if (
        run.status == "requires_action"
        and run.required_action.submit_tool_outputs.tool_calls[0].function.name == "GetTopHeadlinesNews"
    ):
        parameters = json_loads(run.required_action.submit_tool_outputs.tool_calls[0].function.arguments)
        response = GetTopHeadlinesNews(parameters)
        tool_outputs = [
            {
                "tool_call_id": run.required_action.submit_tool_outputs.tool_calls[0].id,
                "output": json_dumps(response),
            }
        ]
        timeStart = time.time()
        run = client.beta.threads.runs.submit_tool_outputs(
            thread_id=thread.id,
            run_id=run.id,
            tool_outputs=tool_outputs,
        )
        print(f"Time taken to submit tool outputs: {time.time()-timeStart}")
        run = wait_on_run(run, thread)
        return get_response(thread)


# Function calling
def GetTopHeadlinesNews(arguments):
    url = "https://newsapi.org/v2/top-headlines"
    parameters = arguments
    response = requests.get(url, params=parameters, headers={"X-Api-Key": api_key_news})
    data = response.json()
    return data


# Waiting in a loop
def wait_on_run(client, run, thread):
    timeStart = time.time()
    print("Waiting for run to complete")
    while run.status == "queued" or run.status == "in_progress":
        run = client.beta.threads.runs.retrieve(
            thread_id=thread.id,
            run_id=run.id,
        )
    print(f"Time taken to wait on run: {time.time()-timeStart}")
    print(f"Run status: {run.status}")
    return run


def create_thread_and_run(client, assistant_id, user_input):
    timeStart = time.time()
    thread = client.beta.threads.create()
    print(f"Time taken to create thread: {time.time()-timeStart}")
    run = submit_message(client, assistant_id, thread, user_input)
    return thread, run


def submit_message(client, assistant_id, thread, user_message):
    timeStart = time.time()
    print(f"Submitting message...")
    client.beta.threads.messages.create(thread_id=thread.id, role="user", content=user_message)
    print(f"Time taken to submit message: {time.time()-timeStart}")
    print(f"Creating run...")
    return client.beta.threads.runs.create(
        thread_id=thread.id,
        assistant_id=assistant_id,
    )

Python version

Python v3.12.7

Library version

openai v1.52.0

Also Im tried with the directly request to OpenAI and I receive the same performance.