Issue with negative API balance - July 2nd 2025

I still receive the 429 error

It was resolved on my end as well. Did OpenAI explain what happened? Any post-mortem? or at least a brief explanation?

The screen theme was black, so a bug crept in and no one spotted it until morning, when the screen theme switched to white.

Same issue here, credit balance restored to normal but still receiving “insufficient_quota” error.

1 Like

Same here.. Credit balance restored but receiving insufficient_quota error

getting the same error. credit balance is normal but getting quota error. “You exceeded your current quota, please check your plan and billing details.”

Oh, big yikes.

Considering OpenAI has shut down for this week, the solution may take some time. Seriously horrible timing.

Anybody who’s hoping for a rapid fix: it may come, but it may be best to use something like Azure as a back-up, and to always have a back-up planned.

(Did someone vibe-code the payment processing system? LOL. The fact that the billing panel has been “corrected” but the internal one hasn’t is seriously concerning.)

Hi everyone – Hannah from the OpenAI support team here. I’m really sorry for the disruption. We’re looking into this now and I’ll have an update for you shortly.

10 Likes

Indeed some is repaired but as far as i can see, the credit balance as it was is not brought back . Only the top-up i did today to secure production minus todays production is in play. The remainder as it was before the error is not. Production works at least but OpenAI is not there yet. :frowning:

This night someone registered in my beta app an tried to manipulate my bots… And at the same time my balance dropped deep into depth with auto recharge off. I really panicked and deleted all my keys, because I wasn’t able to find the abnormal usage. Then my balance came back, I created at least one new api key, and tryed my web app… Error again. I reached my limits… Panic again… Oh god… What a day. But my web app is now 200% saver :sweat_smile:

1 Like

So I wake up to a credit balance of -150 USD on my account.

I never activated auto-renew because our app is still in testing and I didn’t want to risk getting a crazy bill because something went wrong.

We’re building agents you know, that choose THEMSELVES of what, when and how many calls to make.

So that’s why I’d always add credit, let it run out, then add more.

Yet all of a sudden IM NEGATIVE 150 US DOLLARS.

The weird part is my usage reports says I’ve spent like 30 bucks in the last week or month or something.

I suspect the reason to be the following: during the night our agent made a tool call or something and got the error from the API that i had exceeded the quota. Yet, it kept retrying the operation, and this went on all night. I suspect OpenAI is billing me for all these failed requests.

I can’t be offline right now, nor I can afford this stupid bill. It says support should get back to me in 2 days. I can’t wait that long.

Can someone please help?

Seems duplicate with Issue with negative API balance - July 2nd 2025

hey guys i know this has been an on-going issue or not anything new - but ive tried contacting support using their chat on the help page and no one has answered so was really hoping for some guidance on here

i want to connect my chat gpt to my widget for my website

can run my widget locally and even ran a curl command and got back:

Fantastic! Your last curl command worked and returned a list of available models from OpenAI.

This means:

  • Your API key is valid
  • Your account has quota
  • You have access to the latest models, including gpt-4o, gpt-4-turbo, and gpt-3.5-turbo

I am getting an “insufficient_quota” error when using the API, even though my organization has a $39.99 credit balance and my monthly budget is set to $100. I have not hit my usage limits, and my API key is from the correct project. I can list models with the API key, but chat completions return a 429 insufficient_quota error. Can anyone help resolve this

i even type in “hi” to the chatbot and get back the 500 error message

:cross_mark: Error: 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’}}
INFO: 127.0.0.1:56357 - “POST /chat HTTP/1.1” 500 Internal Server Error

please someone help :frowning:

I’m receiving the same error and replying to other posts that match - even did the same and ping the models endpoint successfully. I have a tier 5 and almost $100 in credit. Nowhere near the limits for testing an app sending “hello.”

I have the same problem. Enough money on my (tier 3) account and it stopped processing my requests with the same error…

I have been using codex in terminal lately, is this the reason? Did it consume too much tokens or something?

so what can I do here? Can I throw my account? Is there a way to let openai check this? How can I contact support?

i had the same issue today. i had $40 usd in credits, i started getting insufficient quota. i had my limits in 300 usd per month (used only 120). the solution that worked is adding 100 usd on top of it. wtf is going on? whats this bug? now im super worried that this will happen somehow in the future unexpectedly

This is the second time I faced this issue.
The first time, it resolved on its own after hours of trying and struggling.

Now I face this again.
My usage shows well below the budget
My billing page shows good amount of unspent dollars.

But I still face this issue and I don’t know what to do. I need help asap.

REQUEST END [b63b4ec1] - Unexpected error: OpenAI API quota exceeded. Please add credits to your OpenAI account at https://platform.openai.com/account/billing

Our organization is also seeing 429 errors in the API. How do we report this so that it shows on Status page of openai?

The status page reflects the up-time of their services. Logical errors in the actual code is separate.

There’s no other way to see the current status except for posts here, or on social media.