Anthropic Outage History

Anthropic Claude 3.5 Haiku experienced elevated errors affecting requests across api.anthropic.com, Claude.ai, and the Anthropic Console. Users attempting to use Claude 3.5 Haiku endpoints may have encountered failures or timeouts. The issue began at May 8, 2025 2:08 PM UTC and was resolved by May 8, 2025 3:08 PM UTC, with the outage impacting three components: api.anthropic.com, console.anthropic.com, and claude.ai. The incident prompted an investigation by Anthropic and led to two status updates from the provider during the event.

The Claude 3.5 Haiku service experienced a short interval of elevated errors affecting requests across the API, Claude.ai, and the Anthropic Console. The incident began at May 8, 2025 5:03 AM UTC and was resolved by May 8, 2025 5:18 AM UTC, lasting about 15 minutes. Users attempting interactions with Claude 3.5 Haiku during that window may have encountered transient failures or degraded responses. The issue was communicated via two official status updates describing the investigation and final restoration across the affected components: api.anthropic.com, claude.ai, and console.anthropic.com.

Anthropic Claude experienced a brief degradation affecting three components: console.anthropic.com, api.anthropic.com, and claude.ai. Users attempting Claude 3 Sonnet requests could have encountered errors during the incident window. The outage began on 9:53 PM UTC and was resolved by 10:13 PM UTC.

The provider issued 2 status updates during the incident, noting that the issue was identified and a fix was being implemented, and that services were restored by 10:13 PM UTC.

Anthropic Claude experienced a brief degradation affecting Claude 3.7 Sonnet, with elevated errors impacting api.anthropic.com and claude.ai. The incident caused intermittent failures for some users requesting Claude services. The issue began at May 7, 2025 8:08 PM UTC and was resolved about 35 minutes later at May 7, 2025 8:43 PM UTC.

Anthropic Claude experienced a brief degradation tied to the Claude 3.7 Sonnet deployment. At 6:53 AM UTC the status page reported elevated errors affecting api.anthropic.com and claude.ai, prompting an investigation. The incident lasted until 7:18 AM UTC when services returned to normal.

Two official updates were posted during the incident. The components api.anthropic.com and claude.ai briefly transitioned to a degraded state (warn) before returning to normal once the issue was resolved.

An outage affected Anthropic Claude services for about 40 minutes, impacting Claude AI (claude.ai), the API at api.anthropic.com, and the Console at console.anthropic.com. Users likely experienced errors or degraded performance during the window. The incident was listed as degraded (warn) on the status page with the note Elevated Errors on Haiku 3.5 and an ongoing investigation. The outage began at 2:18 PM UTC and was resolved by 2:58 PM UTC.

The provider published two status updates during the incident and restored all affected components to normal by the end of the window.

Anthropic Claude 3 Opus experienced a brief period of elevated errors affecting API access (api.anthropic.com) and Claude AI (claude.ai). The Claude status page listed an investigation beginning at 11:08 PM UTC and, after about 40 minutes, the services returned to normal, with both components restored. The incident involved two components and was accompanied by two official status updates. The outage began on 11:08 PM UTC and was resolved by 11:48 PM UTC.

Anthropic Claude.ai free tier users on the 3.7 Sonnet model experienced elevated errors beginning at 12:30 PM UTC. Pro and Max plans were unaffected, so the impact was limited to free-tier usage. The incident persisted for nearly 3 hours before recovery.

Claude.ai's status transitioned to degraded at Apr 30, 2025 2:58 PM UTC and returned to normal by 3:28 PM UTC after root-cause mitigation. The provider published two official updates during the incident documenting the elevated errors and recovery.

Anthropic Claude experienced a brief outage with model request failures across all Claude models. This impacted api.anthropic.com, claude.ai, and console.anthropic.com, causing errors for users attempting to run Claude models. The incident lasted about 1 hour and 45 minutes, from 2:23 PM UTC to 4:08 PM UTC. The service issued two status updates during the incident before returning to normal.

An incident on Apr 29, 2025 8:58 AM UTC affected api.anthropic.com, claude.ai, and console.anthropic.com, causing elevated errors and degraded performance for Claude 3.7 Sonnet. The outage lasted until Apr 29, 2025 10:28 AM UTC, roughly 1 hour and 30 minutes, and impacted three components: api.anthropic.com, claude.ai, and console.anthropic.com. Users attempting to access the Claude API, web console, or Claude assistant encountered errors during the affected window.

Anthropic's status page documented an investigation into the elevated errors across the three components and noted two official updates. The incident was described as a repeat of an earlier incident that day, and services were restored after the final update.