Recurring Error message code 429

Hi, I just start a prepaid account. I did not use the free before it is expired.
However every time I try to use the service I receive the same error:
Failed to generate content using AI writer: gpt-3.5-turbo - error: Error: You exceeded your OpenAI quota limit, please wait a period for the quota to refill (chat initial call).
I have $9.96 on the account and my free access is expired.
Is there a solution to the problem.

Error 429 is a rate limit error, you need to slow down your requests.

1 Like

I am trying to write 5 articles once. what do you mean by request?

Welcome to the community!

Requests are messages you are sending to it. So if you are asking it to write 5 articles it is responding a lot to your requests sparking the error message.

Really? It is not so powerful. Then let me try one simple article now and let you know if it works

1 Like

Not in power really, just keep in mind it takes around a quart of water to cool one of their servers about every 50 prompts.

Yes it works. Thanks but the article is too long. I understand why it won’t work for 5 like that. Look at the report below. Does that mean I use too much token. I should work on my prompt also I think.
1:40 PM
Local time: Sep 21, 2023, 9:40 AM

gpt-3.5-turbo-0613, 4 requests

1,686 prompt + 1,963 completion = 3,649 tokens

Yep, if you need any more help just ask. :smiley:

Accounts that don’t have a payment method are limited to three requests per minute as a documented rate limit. Then another limit for the first 48 hours after a payment method has been input, then another limit after 48 hours.

Then new limits available based on approval of request using criteria.

gpt-3.5-turbo limits of a paying mature account of 90,000 tokens per minute should take a decent amount of recurring token requests to reach, or specific patterns to result in a block.

Tokens of a request input are estimated, and your max_tokens considered as its rate consumption.

For example, requesting n=100 for one-hundred runs of the chat completion, with max_tokens of 3000 set, would be blocked by being estimated as a rate of 300,000+ tokens (even if it would run for five minutes of generation).

You can discover what tier your account is in and what is needed to keep your requests per minute below that.