Model output quality

This incident has now been resolved. We've published a technical postmortem on https://anthropic.com/engineering/a-postmortem-of-three-recent-issues.

We've identified the root causes of the reported quality issues and deployed mitigations for each. We do not see any ongoing issues at this time. We are continuing to monitor and evaluate these mitigations out of an abundance of caution.

A technical post-mortem will be published on our engineering blog next week.

The second issue has also been identified to impact Opus 3.

Last week, we opened an incident to investigate degraded quality in some Claude model responses. We found two separate issues that we’ve now resolved. We are continuing to monitor for any ongoing quality issues, including reports of degradation for Claude Opus 4.1.

Resolved issue 1 - A small percentage of Claude Sonnet 4 requests experienced degraded output quality due to a bug from Aug 5-Sep 4, with the impact increasing from Aug 29-Sep 4. A fix has been rolled out and this incident has been resolved.

Resolved issue 2 - A separate bug affected output quality for some Claude Haiku 3.5 and Claude Sonnet 4 requests from Aug 26-Sep 5. A fix has been rolled out and this incident has been resolved.

Importantly, we never intentionally degrade model quality as a result of demand or other factors, and the issues mentioned above stem from unrelated bugs.

We're grateful to the detailed community reports that helped us identify and isolate these bugs. We're continuing to investigate and will share an update by the end of the week.