Time within recorded incidents — October 2026
October 2026 is still running. These are month-to-date figures measured from October 1, 2026 up to now, derived from SpaceXAI's public incident reporting, and they will keep moving until the month closes.
Month to date — not final. Locked when October 2026 ends.
By component
- Grok on the web0.0%0 · 0 min
- Grok in X0.0%0 · 0 min
- Grok for iOS0.0%0 · 0 min
- Grok for Android0.0%0 · 0 min
- Grok Build0.0%0 · 0 min
- Office and Workspace plugins0.0%0 · 0 min
- Single sign-on0.0%0 · 0 min
- API — us-east-10.0%0 · 0 min
- API — eu-west-10.0%0 · 0 min
- API console0.0%0 · 0 min
Share of October 2026 inside a recorded incident window, incidents recorded, and total incident time.
How this is counted
- Source. SpaceXAI's public incident reporting for Grok. 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. The window measured here is 427 h 14 min.
- 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.