Back to Now
Claude tea Developing

Claude hit elevated errors across three models

Anthropic reported elevated errors for Mythos 5.1, Fable 5.1 and Opus 5 before success rates returned to normal.

Why now The status page moved from investigation to monitoring, making this a live reliability event rather than a retrospective complaint. 3 models affected

What happened

Anthropic’s status page reported elevated request errors affecting Claude Mythos 5.1, Claude Fable 5.1 and Claude Opus 5. The company later said success rates had returned to normal and moved the incident into monitoring.

The incident reached 95 points on Hacker News in the first public discussion snapshot, reflecting how quickly provider reliability becomes developer news when coding and agent workflows depend on it.

Why it matters

Switching from one Claude tier to another is not a complete resilience plan if several models share the same failure domain. Production agents need provider-level fallbacks, retry budgets and a way to stop repeated side effects when a request outcome is uncertain.

What remains unknown

The public status update did not yet provide a root cause. Duration, regional impact and any final incident report remain the next useful pieces of evidence.