Claude suffers brief multi-model outage
Anthropic resolved a major service disruption that caused elevated error rates across multiple Claude models for approximately 85 minutes. The incident affected the Claude API, Claude Console, Claude.ai, Claude Code, and Claude Cowork before a fix restored normal service levels.
While brief, this outage highlights the fragile dependency developers face when building production applications on a single AI provider's API. For tools like Claude Code and agentic workflows, even minor service degradation immediately halts active development.
- –The incident lasted from 14:08 to 15:33 UTC, affecting all core entry points including the API, Console, and developer CLI tools.
- –Developers experienced 'server overloaded' errors and model unavailability, highlighting the need for local fallback models or multi-provider API redundancy.
- –This disruption occurred shortly after the introduction of Claude Code and Cowork, emphasizing the high stakes of reliability as Anthropic pushes deeper into active agentic developer tooling.
- –Anthropic identified and mitigated the issue within 17 minutes of initiating the investigation, demonstrating swift incident response despite the broad surface area of the failure.
DISCOVERED
1h ago
2026-06-23
PUBLISHED
3h ago
2026-06-23
RELEVANCE
AUTHOR
rob