Errors with the Fable 5 Model in Copilot
This summary is created by Generative AI and may differ from the actual content.
Overview
On August 13th, 2026, between approximately 14:06 and 15:47 UTC, the Copilot service experienced degradation of the Claude Fable 5 model due to an issue with an upstream provider. Users encountered elevated error rates, peaking at 43% and averaging 12%. Users who selected Auto or alternative models were unaffected. The issue was resolved by mitigation put in place by the upstream provider.
Impact
Elevated error rates for Fable 5 model users, peaking at 43% and averaging 12% during the 101-minute incident window. Users selecting Auto or alternative models experienced no impact.
Trigger
Issue with upstream model provider
Detection
Reports of degraded performance for Copilot AI Model Providers were investigated starting at 14:43 UTC. Degraded availability was confirmed and communicated to users at 14:50 UTC.
Resolution
Mitigation was put in place by the upstream provider, resolving the issue by 15:47 UTC. GitHub recommended users select alternative models or Auto as a workaround during the incident. GitHub is working with the upstream provider to further improve service resiliency to prevent similar incidents in the future.
Root Cause
Issue with upstream model provider. Specific technical details were not disclosed in the postmortem.