RateLimitError: 429 with gpt-4 and gpt-3.5-turbo

I am a paying customer, and I am using an older version of OpenAI (openai-0.28.0). I am facing an issue with Ratelimiterror:

RateLimitError: 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.

I used all the advice available online and checked out the page for model limits and error handling. I found a code snippet for verifying the error code, and here is the output:

openai.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’}}

Everything was working fine till the 29th of December, but on the 31st, when I ran the code, it started throwing errors, and now it’s been 4 days.

I added another $10 to my account just in case the available credit of $9 wasn’t enough, but even that didn’t solve the issue. I am using the API in the simplest form possible (I took the code from the OpenAI website). I have tried everything, but nothing seems to be working. Can anyone help me, please? I guess I can create a new account, but I would lose $20 worth of credit.

Thank you very much!

Have you done a pip install --upgrade openai ? if not is there a reason for that?

1 Like

Thanks @Foxalabs! Yeah I already tried that. Its not the version which is the problem I guess.

What is your monthly budget limit set at?

Also can you post a code snippet of how you are making the API calls?

Hi! I have a quick troubleshooting step that might have been overlooked.

First, get the OPENAI_API_KEY that you’ve set in the environment of your operating system. This is used by default in the new client without you needing to specify.

Then when you go to your account, give a single-click to your name icon in the lower left, and a menu will appear. See if you have two organization, with one being a “personal” account that was added automatically to your older account with named organization.

If you have two selectable organizations, there’s a distinct possibility that API keys were obtained from the wrong one, or that payments were made to fund the wrong one.

Try the organization profile that’s not personal. See if that’s the one that both contains your currently used API key (by the last digits), and also shows a credit balance or that you have a monthly billing account. After you are sure you have the correct organization, then proceed to API keys and simply make yourself a new key to spend the available funds.

There have been delays reported in being able to spend the funds of recently-added payments. This was also seen last month at the start of the month.

1 Like

maybe we should Upgrad the openai library for python