Degradation with Gemini 3.8 Flash

Summary

Degradation with Gemini 3.8 Flash

On September 16, 2026, between 04:40 and 11:45 UTC, the Gemini 3.8 Flash model in GitHub Copilot experienced degraded availability. Requests to this model failed at an average rate of 6.4%, and the impact was highest during peak traffic hours. Other Copilot models were not affected. Users could continue to work with a different model or with 'Auto'.<br /><br />The degradation was caused due to a capacity issue with an upstream model provider. Failure rates returned to normal as the daily traffic peak passed. We monitored the model until it was healthy and resolved the incident at 17:48 UTC. We are working to make our systems resilient to cover peak demand for all the Copilot models.

• Started

  Sep 16, 2026 · 7:20 AM

• Resolved

  Sep 16, 2026 · 5:48 PM

• Duration

  10h 27m

• Severity

  Major

• ◐

  Investigating

  Sep 16 · 7:21 AM GitHub

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

• ◐

  Investigating

  Sep 16 · 8:28 AM GitHub

  Copilot AI Model Providers is experiencing degraded performance. We are continuing to investigate.

• ◐

  Investigating

  Sep 16 · 10:58 AM GitHub

  Copilot AI Model Providers is experiencing degraded performance. We are continuing to investigate.

• ◉

  Monitoring

  Sep 16 · 11:45 AM GitHub

  The degradation affecting Copilot AI Model Providers has been mitigated. We are monitoring to ensure stability.

• ✓

  Resolved

  Sep 16 · 5:48 PM GitHub

  On September 16, 2026, between 04:40 and 11:45 UTC, the Gemini 3.8 Flash model in GitHub Copilot experienced degraded availability. Requests to this model failed at an average rate of 6.4%, and the impact was highest during peak traffic hours. Other Copilot models were not affected. Users could continue to work with a different model or with 'Auto'.<br /><br />The degradation was caused due to a capacity issue with an upstream model provider. Failure rates returned to normal as the daily traffic peak passed. We monitored the model until it was healthy and resolved the incident at 17:48 UTC. We are working to make our systems resilient to cover peak demand for all the Copilot models.