API access to GPT-4 no longer working

I am working on an application which uses GPT-4 API calls. It was working last night, but as of this morning all of my API calls are failing. Here’s some example Python code for testing:

from openai import OpenAI

LLM = OpenAI()

response = LLM.chat.completions.create(
    messages=[{'role': 'user', 'content': 'What is 1+1?'}],

(My API token is stored in the environment variable OPENAI_API_KEY.)

The response I get is:

NotFoundError: Error code: 404 - {'error': {'message': 'The model `gpt-4` does not exist or you do not have access to it. Learn more: https://help.openai.com/en/articles/7102672-how-can-i-access-gpt-4.', 'type': 'invalid_request_error', 'param': None, 'code': 'model_not_found'}}

Interestingly, if I use gpt-3.5-turbo instead of gpt-4, then I get the error:

RateLimitError: Error code: 429 - {'error': {'message': 'You exceeded your current quota, please check your plan and billing details. For more information on this error, read the docs: https://platform.openai.com/docs/guides/error-codes/api-errors.', 'type': 'insufficient_quota', 'param': None, 'code': 'insufficient_quota'}}

I have confirmed in the admin UI that I am nowhere near any caps for this model.

Anyone experiencing something similar?

Check, re-create, and update your API key.

I generated a new API key and got the same result. I also tried an API key associated to a different ChatGPT Plus account and got the same result.

Go to the Playground

Try and use the the GPT-4 model. If it works then in the left side-bar click API keys and make sure the API key there is the same API key in your env

It does work in the Playground, and as far as I can tell it’s using the same key as the one I’m using on my local machine. Does that mean it’s some sort of config issue on my local machine?

Try and hardcode the API key to check

from openai import OpenAI

LLM = OpenAI(api_key=[pasted_key])

OK, I have to run to a meeting, but I’ll try after - thanks for he help so far!

What Tier for billing are you?

Are you aware of the rate limits? Have enough credits in your account?

Update: after I got back from my meetings yesterday I generated a new API key and everything worked fine. I definitely also tried that yesterday morning, so I’m not sure what the problem was, but :person_shrugging:

The GPT-4 model still not working April 2024