N8n → OpenAI API suddenly failing (500 error) — used to work, now all calls fail

API Bugs
1

Hi everyone!

I have an n8n workflow that calls the OpenAI “Message a Model” API (GPT-4o-mini) to generate 10 search keywords.

This node has worked for weeks with no changes, but starting today it fails every single time.

Error I’m getting

Error code: 500 (server_error)

Full message:
{
  "error": {
    "message": "An error occurred while processing your request. You can retry your request, or contact us through our help center at help.openai.com if the error persists. Please include the request ID req_08ea169a02c145bc8c65dbe1d4e3476c in your message.",
    "type": "server_error",
    "param": null,
    "code": "server_error"
  }
}

This happens every time, and I see no logs in my OpenAI usage dashboard, even though the API key shows activity (“last used today”).

What I already checked

  • API key is correct

  • Billing is active

  • No rate limits

  • Other API tools (curl) work fine

  • n8n node config is unchanged

  • Same workflow worked yesterday

  • Restarted workflow, recreated OpenAI credentials, still failing

  • Tested with GPT-4o-mini, GPT-4o, and GPT-4 — all fail

  • Removed expressions from prompt — still fails

  • Calls never appear in my OpenAI request logs

Is there a known outage, API regression, or request schema change affecting Message a Model (Responses API) calls?

Why would requests suddenly return 500 and not show up in the usage logs at all?

Here are two failing request IDs:

  • req_7228f8c960dc437bbdffe62e9a6ff251

  • req_08ea169a02c145bc8c65dbe1d4e3476c

Any guidance is deeply appreciated — this completely broke my production workflow.

1 Like