Anthropic's Claude Sonnet 5.5 saw two separate bouts of elevated errors on 27 October, lasting 45 minutes and 42 minutes. Both affected claude.ai and api.anthropic.com, and both were Anthropic's fault, not yours.
Started
09:14 UTC
Duration
Lasted 45m
Source
IsDown
Next time
Anthropic is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
Affected components: claude.ai, api.anthropic.com
What happened?
Anthropic logged two incidents on 27 October, both titled "Elevated errors for Claude Sonnet 5.5". The first started at 09:14 UTC. Anthropic said it was investigating elevated errors on requests to Claude Sonnet 5.5, deployed a fix by 09:38, and marked it resolved at 09:56, 45 minutes after it began. The second started at 15:09 UTC and was marked "identified" straight away rather than "investigating", suggesting Anthropic already had a lead on the cause. That one closed at 15:50, 42 minutes later, and Anthropic's resolution note gave a more precise window: impact from 07:42 to 08:17 Pacific, or 14:42 to 15:17 UTC, slightly narrower than the full incident window. Both incidents named claude.ai and api.anthropic.com, and the second resolution note also listed Claude Code and Claude Cowork as affected. Anthropic did not publish a cause for either incident beyond "elevated errors", and no root cause has been filed as of the last update. No user reports were filed in this window, which is not unusual for errors concentrated on a single model rather than a platform-wide failure. History shows this is a recurring pattern for Anthropic: 54 incidents in the last 90 days, with a median duration of 47 minutes, so these two sit close to Anthropic's own typical outage length. Two incidents on one day, both tied to the same model and roughly the same duration, point to a repeating failure mode rather than a one off.
Learning
Over the last 30 days, UptimeRobot's own checks against Anthropic logged 15 incidents, all resolved, with a median duration of 42 minutes, a figure close to what Anthropic itself reports for this day's incidents. That overlap matters: it means an external check watching Anthropic's actual endpoints caught the same shape of failure that Anthropic's own status page describes, rather than missing it or disagreeing with it. If your own monitoring only watches your product's front end, elevated errors on a specific model like Claude Sonnet 5.5 can slip through unnoticed until your users start filing reports, because your endpoint might still return a 200 while the underlying model call fails or slows. That is the general shape of this kind of outage: the fault sits inside the provider's inference layer, not in anything you control, so watching your own uptime tells you nothing about whether Anthropic is having a bad day. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.