Assistant API QuickStart Error

API
1

Im new to the Openai API and I am trying to experiment with assistants. I have installed the SDK and stored my API Key as a system variable. When following the steps in the QuickStart I am running into the error of not getting a response from my assistant and the output simply just says failed. Here is my code.

from openai import OpenAI

client = OpenAI()

assistant = client.beta.assistants.create(
    name="Math Tutor",
    instructions="You are a personal math tutor. Write and run code to answer math questions.",
    tools=[{"type": "code_interpreter"}],
    model="gpt-4o-mini",
)

thread = client.beta.threads.create()

message = client.beta.threads.messages.create(
  thread_id=thread.id,
  role="user",
  content="I need to solve the equation `3x + 11 = 14`. Can you help me?"
)

run = client.beta.threads.runs.create_and_poll(
  thread_id=thread.id,
  assistant_id=assistant.id,
  instructions="Please address the user as Jane Doe. The user has a premium account."
)

if run.status == 'completed':
  messages = client.beta.threads.messages.list(
    thread_id=thread.id
  )
  print(messages)
else:
  print(run.status)
1 Like
2

I’m getting the same, a “failed” status that just says LastError(code='server_error', message='Sorry, something went wrong.')
status.openai.com doesn’t have any info on gpt-4o-mini not working.
I’m also using client.beta.threads.create_and_run() since that’s mentioned in the API docs and I’m using tools.
I would love some clarification here on what’s best to use. I do my own polling with a handler for when there is a 'requires_action' in order to submit my tool outputs.
I also have another assistant with tools on 'gpt-4o' (not mini) I’ve been using that has been working fine, although I’m not using the create_and_run() method, I’m doing that manually. I’ll try that with my mini assistant and report back.

1 Like
3

My responses are still failing. Same lack of reason given, the status page should reflect this. For all intents and purposes gpt-4o-mini is down. It seems to work on the playground, not sure what’s special about my API calls.

Edit 08-03:
I switched to using gpt-3.5-turbo (the only change I made) and now it’s working. So very clearly there’s an issue with gpt-4o-mini which should be reflected in the status page.

2 Likes