Anthropic resolves elevated errors affecting five Claude models
Anthropic resolved an incident that caused elevated error rates across five Claude models; the official record says impact lasted from 06:05 to 06:35 UTC on October 25.
AT A GLANCE
Key points
- The incident affected five Claude models and was marked resolved at 07:53 UTC on October 25.
- Anthropic said user impact lasted from 06:05 to 06:35 UTC.
- Claude services are again listed as operational on the official status page.
What is changing?
Anthropic investigated elevated error rates on requests to Claude Mythos 5.1, Fable 5.1, Opus 5.5, Sonnet 5.5 and Haiku 5.5 on October 25. The first notice on the company’s official status page was posted at 06:14 UTC.
Anthropic said it identified the cause at 06:27 UTC and deployed a fix at 06:35 UTC. The incident was marked resolved at 07:53 UTC. According to the official record, user impact lasted for 30 minutes, from 06:05 to 06:35 UTC.
The status page currently shows claude.ai, Claude Console, the Claude API, Claude Code, Claude Cowork and Claude for Government as operational. The incident concerned temporary service error rates, not a change in model quality.
Explore the original sources
Source date: 25 October 2026.
Published by AI Radar: 25 Oct 2026, 09:59 (Istanbul, UTC+3). Updated: 25 Oct 2026, 12:01.
FOLLOW THE TOPIC
Topics and tags
Share your discovery.
Send this story to a friend.
Reader comments
What do you think about this development? Share your experience or question. Comments are published after editorial approval.
Your name/nickname and comment become public if approved. Read: Privacy notice.
Loading comments…