Incident with Copilot AI Model Providers

Summary

Incident with Copilot AI Model Providers

On July 29, 2026, between 19:45 UTC and 21:51 UTC, users of the Fable 5 model in GitHub Copilot experienced increased request failures and latency. The average failure rate across all Copilot requests was 0.006%, while failures for Fable 5 peaked at 21%. Other models remained available. This was caused by degradation of an upstream model provider.<br /><br />The affected endpoint recovered, and we monitored the service until error rates and latency returned to normal levels. We are working to add endpoint redundancy to mitigate similar provider issues in the future.

• Started

  Jul 29, 2026 · 8:07 PM

• Resolved

  Jul 29, 2026 · 9:51 PM

• Duration

  1h 44m

• Severity

  Minor

• ◐

  Investigating

  Jul 29 · 8:07 PM GitHub

  We are investigating reports of degraded performance for Copilot AI Model Providers

• ◐

  Investigating

  Jul 29 · 8:07 PM GitHub

  We are seeing increased error rates with requests to specific model providers.

• ◐

  Investigating

  Jul 29 · 8:18 PM GitHub

  We are investigating increased error rates affecting GitHub Copilot requests to external AI model providers. Some users may experience failures or degraded performance when using Copilot features.

• ◐

  Investigating

  Jul 29 · 8:38 PM GitHub

  The external AI model provider has identified the issue and is working to resolve.

• ◐

  Investigating

  Jul 29 · 9:08 PM GitHub

  The external AI model provider is continuing to investigate.

• ◐

  Investigating

  Jul 29 · 9:51 PM GitHub

  The external ai model provider has resolved the issues, and we have verified Copilot's traffic is fully recovered.

• ✓

  Resolved

  Jul 29 · 9:51 PM GitHub

  On July 29, 2026, between 19:45 UTC and 21:51 UTC, users of the Fable 5 model in GitHub Copilot experienced increased request failures and latency. The average failure rate across all Copilot requests was 0.006%, while failures for Fable 5 peaked at 21%. Other models remained available. This was caused by degradation of an upstream model provider.<br /><br />The affected endpoint recovered, and we monitored the service until error rates and latency returned to normal levels. We are working to add endpoint redundancy to mitigate similar provider issues in the future.