Elevated error rate on Claude Opus 4.8
Claude experienced an elevated error rate on Opus 4.8, starting at 20:42 UTC on July 11, 2025. The issue was identified and a fix was implemented, with services returning to normal operation.
Background
- Anthropic is the AI company behind Claude, a family of large language models. This incident page tracks real-time service disruptions.
- Claude Opus 4.8 is a premium, high-performance version of the Claude model, typically available to paying subscribers via the claude.ai website and API.
- This status page entry reports that users of Opus 4.8 may see increased errors (e.g., failed message generation, timeouts, or "something went wrong" messages). The incident is under investigation — no root cause or fix timeline has been posted yet.
- For context: AI model providers like Anthropic, OpenAI, and Google publish separate status pages for their API and consumer products, so paying customers can track outages separately from general product news.