Claude Console incident history
The last recorded incident affecting Claude Console started on September 1, 2026. Every figure on this page is computed from our archive of Anthropic's public incident reporting.
- Incidents (archive)
- 23
- Last 30 days
- 2
- Median duration
- 1 h 10 min
- Total incident time
- 34 h 49 min
By month
- September 20262 · 3 h
- August 20265 · 3 h 4 min
- July 202616 · 28 h 45 min
Incidents affecting Claude Console
- September 1, 2026Delays in credit purchasesMinor · 1 h 58 min
- September 1, 2026Degraded performance on platform.claude.com and Claude for Microsoft Office 365Minor · 1 h 2 min
- August 16, 2026Service disruption on Claude servicesMajor · Ongoing
- August 16, 2026Service disruption on Claude servicesCritical · 36 min
- August 4, 2026Elevated errors on Claude Sonnet 5Minor · 25 min
- August 4, 2026Elevated errors on Claude Sonnet 5Minor · 39 min
- August 3, 2026Error rates across multiple modelsMinor · 1 h 24 min
- July 26, 2026Elevated errors for Opus 5Major · 1 h 27 min
- July 24, 2026Elevated error rate on Opus 4.8Minor · 1 h 12 min
- July 22, 2026Elevated error rates for Sonnet 5Minor · 48 min
- July 22, 2026Elevated errors on Opus 4.8 and Haiku 4.5Minor · 1 h 46 min
- July 21, 2026Elevated errors on Haiku 4.5Major · 3 h 34 min
- July 21, 2026Elevated errors across Fable 5Minor · 48 min
- July 21, 2026Elevated errors on Haiku 4.5Minor · 1 h 41 min
- July 20, 2026Elevated errors for Claude Opus 4.8Minor · 1 h 35 min
- July 20, 2026Elevated errors on Haiku 4.5Minor · 2 h 20 min
- July 20, 2026Elevated error rates for Opus 4.5Minor · 5 h 44 min
- July 17, 2026Elevated error rates for Sonnet 5Major · 15 min
- July 17, 2026Elevated errors for Claude Opus 4.8Minor · 36 min
- July 16, 2026Enterprise SSO sign-in failuresMajor · 1 h 4 min
- July 16, 2026Elevated errors for Claude Opus 4.7Minor · 4 h 33 min
- July 16, 2026Elevated errors on Claude Sonnet 5Minor · 15 min
- July 15, 2026Elevated errors on multiple modelsMinor · 1 h 7 min
How this is counted
- Source. Anthropic's public incident reporting for Claude. We do not probe, ping or measure the services ourselves, so nothing here is an availability measurement.
- What we count. Minutes of the month that fell inside a recorded incident window for that component — from the reported start to the reported resolution, clipped to the month.
- What we do not count. Severity is ignored: an incident logged as degraded performance counts the same as a full outage. Unreported disruption, regional differences and per-user impact are invisible to us. An incident window is not the same thing as downtime.
- Overlaps. Figures are per component only, never aggregated into one service-level number. Within a component, overlapping incidents count the shared minutes once; the same minute can appear under several components when an incident affected several.
Computed from our incident archive, which starts on — 63 days deep.