OpenAI API shows elevated errors on gpt-6 Luna, mitigation applied

Chat Completions and Responses are degraded, and the company says it is monitoring recovery.

Share

Published 1 min read

Cover for the story about the OpenAI incident
Image: Arte: jkit.tools

In brief

  • OpenAI's API has had elevated error rates on gpt-6 Luna requests since 4:01 PM UTC on November 19, 2026.
  • Chat Completions and Responses are both degraded.
  • The status page lists the incident as monitoring after a mitigation was applied.

OpenAI is seeing elevated error rates on gpt-6 Luna API requests, a problem that began at 4:01 PM UTC on November 19, 2026, according to the official status page. The incident is in the monitoring state.

Two endpoints are listed as degraded: Chat Completions and Responses.

The first update, posted at 4:01 PM UTC, said OpenAI was investigating a potential issue and checking how it affected users. At 4:21 PM UTC, the company said it had applied a mitigation and was monitoring the affected services to confirm recovery.

Live on Radar

Service status now: Discord, GitHub, OpenAI, Roblox and more

See live service status and the incident history on the Radar.

Share

How this story was made

Drafted with the help of artificial intelligence from a fact sheet built on the sources listed below. Software checked its numbers, dates and names against that fact sheet, and Vynuo, the editor in charge, approved the text before it was published.

Sources for this story

How we know: information from an official source (OpenAI Status), checked on November 19, 2026.

  1. OfficialOpenAI: status page (opens in a new tab)OpenAI Status, accessed November 19, 2026
  2. DataStatus OpenAI (opens in a new tab)OpenAI Status, accessed November 19, 2026

Topics

More in AI