Claude model API outage lasting over two hours is resolved
On 18 August 2026, Anthropic addressed a Claude API outage involving increased error rates and degraded performance across multiple Claude models; the impact lasted 16:11–18:23 UTC and was resolved at 19:01 UTC.
On 18 August 2026, Anthropic reported an incident involving degraded performance affecting multiple Claude models on the Claude Status page. According to the updates, the investigation began at 16:20 UTC, when the company stated that it was investigating increased error rates for Claude Mythos 5, Claude Fable 5, Claude Opus 5, Claude Sonnet 5, Claude Haiku 4.5 and other Claude models.
During the incident, the scope narrowed in subsequent updates: the report at 17:12 UTC referred exclusively to increased error rates for Claude Opus 5. At 18:26 UTC, the company announced that it had deployed a fix and was monitoring the results. The source provides no explanation for this change in scope between the initial report (five or more models) and later updates (only Claude Opus 5).
At 19:01 UTC, Anthropic announced that the problem with Claude Opus 5 had been resolved, stating that the impact lasted from 16:11 to 18:23 UTC, approximately two hours and twelve minutes. The source does not provide details about the technical cause of the incident.
Why it matters
Anyone who integrated Claude API into their own application or business process may have encountered error responses or slowdowns during the given two-hour window, which may have affected end-user services built on these models. The incident is closed, so the main practical step is to check retrospectively whether the outage occurred at a critical point in your own operations.
Two audiences, two different impacts
What this means
For individuals
Anyone using Claude API during the period 16:11–18:23 UTC on 18 August 2026 may have experienced request errors or slower responses; the problem is now resolved and no further action is necessary.
For a business
Companies running production systems on Claude API may have experienced increased error rates and reduced call performance for approximately two hours, which may have affected their own customer-facing services; the incident is closed and requires no further intervention apart from a possible impact review.
Risks and complianceCheck the original
Event sources
clearly official source · 1 publisher, 0 independent. We count feeds from the same owner only once.