‹ Back to incident history

Degradation with Gemini 3.8 Flash

ResolvedMajor outageGitHub
Wednesday, September 16, 2026 15:20 – Thursday, September 17, 2026 01:48 (10 hr 27 min)
Affected components
Copilot AI Model Providers
Updates
ResolvedThursday, September 17, 2026 01:48

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.

MonitoringWednesday, September 16, 2026 19:45

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

InvestigatingWednesday, September 16, 2026 18:58

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

InvestigatingWednesday, September 16, 2026 16:28

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

InvestigatingWednesday, September 16, 2026 15:21

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

Incident data from the GitHub official status page:Read the official incident

About this data

Where does the incident data for GitHub come from?

Data comes from the GitHub official status page, syncs every 5 minutes, and covers 90 days. The detail page links to the original report.

How do I read the component status bars?

Each block is one day. Green means no issue was reported, yellow means degraded performance, red means a major outage, and gray means no data. Availability covers the last 90 days.

Why is history missing for some services?

Only services with a public official status page are included.

How is duration calculated?

Duration runs from the official start time to recovery, or to the current time for ongoing incidents. Incidents spanning multiple days appear on each affected day.

See the full service list and live status in the status overview