[gpt-5-nano issue] Something is seriously wrong: "The server had an error while processing your request" + "Your request rate has increased suddenly" errors

API Bugs
1

My entire app is down. We started getting these errors out of nowhere. Nothing works!

“error”: {
“message”: “The server had an error while processing your request. Sorry about that!”,
“type”: “server_error”,
“param”: null,
“code”: null

EDITED:

Now i’m receiving the following error:

AgentFollowUpWorkflow: {
“error”: {
“message”: “Your request rate has increased suddenly, overloading the model. Please reduce your request rate to its original level, keep it stable for at least 15 minutes, and then gradually increase it. If the error persists, please contact us through our help center at help.openai.com.”,
“type”: “slow_down”,
“param”: null,
“code”: “slow_down”
}
}

Note I’m literally not doing absolutely any requests.

EDIT2:

I believe this is happening only with gpt-5-nano

2

Same here!
I just sent 3 messages on the Chat Dashboard for gpt-5-nano, got this error and now every message I send I get that same error.

3

image
image1632×438 20.1 KB

4

Same here:
0|ai-dmx | type: ‘error’,
0|ai-dmx | sequence_number: 2,
0|ai-dmx | error: {
0|ai-dmx | type: ‘slow_down’,
0|ai-dmx | code: ‘slow_down’,
0|ai-dmx | message: ‘Your request rate has increased suddenly, overloading the model. Please reduce your request rate to its original level, keep it stable for at least 15 minutes, and then gradually increase it. If the error persists, please contact us through our help center at help.openai.com.’,
0|ai-dmx | param: null
0|ai-dmx | }
0|ai-dmx | }

5

I’m also using gpt-5-nano

wiht me starte at

| Date Time (UTC) | Sep 25, 2025, 6:49:36 PM | | |

| |
|----|----|----|----|----|----|

ISO 2025-09-25T18:49:36.330196111Z

just did a test , error message is not showing for now

6

This may be related to the currently reported API outage.

7

The error itself is the behavior of the priority tier SLA that is a substitute for prepurchased tiered pricing of compute units. Those requests that grow faster than provisioning should fall back to a normal request instead of you getting no service. Quite odd.

8

I’m sure, I also just check status page, it does have an incident report ongoing.

Read some posts from 2023 and 2024 with same error code, some guys speculated with server being overrun by demand

who knows what’s the issue?

9

Are you talking about the limits by Tier?

10

I’m talking about ramp rate limits for Latency SLA, which is for Enterprise customers under contract.

https://openai.com/api-priority-processing/

You’ll see the error message language being reported directly aligns with that service’s growth limits.

11

Interesting, did not know about this.

Make sense.

Thanks for the info.

12

Issue Resolved.

image
image779×801 44 KB

13

It’s wise not to rely solely on OpenAI. You could back it up with Azure or AWS, or even integrate another model provider entirely. Service outages can happen not just with OpenAI, but with any provider. Having a fallback ensures you get a response from somewhere rather than no response at all! :slight_smile: