ERROR CODES / ENGLISH

OpenAI 500 server error: retry and check the status page

What the OpenAI 500 server error means, when retrying is enough, and how to tell a provider outage from a problem in your own request.

Reviewed 2026-11-13 · ERROR CODES / 500 · Chinese detail: 完整中文页

What it means

The OpenAI server hit an internal error. It is not caused by your request. Retry.

{"error":{"message":"The server had an error processing your request.","type":"server_error"}}

Common causes

ShareCause
40%Configuration problem (key, base URL, variable name)
30%Provider status (rate limiting, overload, outage)
20%Request content (parameters, structure, length)
10%Network path (DNS, proxy, certificate)

The message text names the server, so the provider-status row is the natural first suspect. The generic ranking in the table also lists configuration first; keep it as a checklist if retries do not clear the error.

How to confirm

Wait 10 seconds and retry the identical request. If the same payload succeeds, the failure was transient and on the server side. If it fails again, try a different model to see whether the failure follows the model or the whole account. Then check status.openai.com for an active incident before changing your own code.

Fix

  1. Wait 10 seconds and retry.
  2. Try another model.
  3. Check status.openai.com.

If the status page reports an incident, wait for it to be resolved rather than resending aggressively.

If it still fails

SymptomResponsible party
Still fails after changing the request contentConfiguration or provider
Switching site or network fixes itThat side
Only triggered under specific conditionsYour request

If every retry fails while the status page shows no incident, and another model works, the problem is specific to that model or your access to it. If no model works, move to the configuration and network rows.

General guidance, not a statement about any specific provider. For measured data, see the relay directory and method.