Disruption with OpenAI Models

Summary

Disruption with OpenAI Models

On May 28th, 2026, between approximately 18:27 and 20:41 UTC, the GitHub Copilot service was degraded due to an issue with the Responses API of an upstream provider affecting the GPT-5.2, GPT-5.3-Codex, GPT-5.4, and GPT-5.5 models. Requests routed to these models via the Responses API returned elevated error rates, which also affected Copilot coding agent and Copilot code review. No other models were impacted. <br /><br />We mitigated the incident by shifting traffic away from the affected models while the upstream provider deployed a fix. <br /><br />GitHub is working to improve automated failover for the affected models and strengthen monitoring to prevent similar incidents in the future.

• Started

  May 28, 2026 · 7:01 PM

• Resolved

  May 28, 2026 · 8:41 PM

• Duration

  1h 40m

• Severity

  Critical

• ◐

  Investigating

  May 28 · 7:01 PM GitHub

  We are investigating reports of impacted performance for some GitHub services.

• ◐

  Investigating

  May 28 · 7:20 PM GitHub

  Copilot is experiencing degraded performance. We are continuing to investigate.

• ◐

  Investigating

  May 28 · 7:40 PM GitHub

  We are investigating errors with Copilot requests using OpenAI models

• ◐

  Investigating

  May 28 · 8:06 PM GitHub

  Open AI models are currently unavailable. We are shifting requests to other models to reduce impact.

• ✓

  Resolved

  May 28 · 8:41 PM GitHub

  On May 28th, 2026, between approximately 18:27 and 20:41 UTC, the GitHub Copilot service was degraded due to an issue with the Responses API of an upstream provider affecting the GPT-5.2, GPT-5.3-Codex, GPT-5.4, and GPT-5.5 models. Requests routed to these models via the Responses API returned elevated error rates, which also affected Copilot coding agent and Copilot code review. No other models were impacted. <br /><br />We mitigated the incident by shifting traffic away from the affected models while the upstream provider deployed a fix. <br /><br />GitHub is working to improve automated failover for the affected models and strengthen monitoring to prevent similar incidents in the future.