Anthropic Outage History

Anthropic Claude 3 Haiku experienced degraded performance starting at Apr 16, 2025 4:03 PM UTC with elevated errors affecting api.anthropic.com, claude.ai, and console.anthropic.com. The service issued 2 status updates while engineers investigated the issue. The problems were resolved by Apr 16, 2025 5:08 PM UTC, after about 1 hour and 5 minutes of degraded operation across the Claude endpoints.

Anthropic Claude experienced elevated errors when processing requests to several models, including Claude 3.7 and 3.5 Sonnet, resulting in degraded responses for users calling Claude via api.anthropic.com, claude.ai, and console.anthropic.com. The incident began at 11:28 PM UTC and involved three components that were affected during the period: api.anthropic.com, claude.ai, and console.anthropic.com.

Anthropic Claude.ai experienced a brief authentication issue causing a login outage for users. The incident affected the claude.ai component, with some users unable to sign in. The outage began at Apr 15, 2025 6:13 PM UTC and was resolved by Apr 15, 2025 6:18 PM UTC, lasting about 5 minutes. The provider posted two status updates during the event, noting investigation into login problems and later confirming resolution. The issue appeared isolated to Claude.ai and did not indicate broader service disruption.

The Anthropic Claude API at api.anthropic.com and the Claude.ai frontend experienced a brief partial outage, with users potentially seeing degraded performance or intermittent errors. The incident began at Apr 14, 2025 11:58 PM UTC and was resolved by Apr 15, 2025 12:13 AM UTC, lasting about 15 minutes. The provider listed Sonnet 3-7 as partially degraded and posted two updates during the incident as they worked to implement a fix. Affected components included api.anthropic.com and claude.ai.

Anthropic Claude AI experienced a brief degraded performance incident affecting Claude 3.7 Sonnet. The issue began at 7:48 AM UTC when the claude.ai component entered a degraded state, potentially causing errors or slower responses for users. The status page noted that the issue had been identified and a fix was being implemented, and the degradation was resolved by 8:18 AM UTC, with claude.ai returning to normal. The incident involved two official status updates during its course. User reports began around 7:45 AM UTC, indicating early visibility before official acknowledgment. The outage lasted roughly 30 minutes and primarily impacted the claude.ai component.

The Claude 3.5 Haiku model with vision experienced an outage that prevented image-containing requests from receiving responses about image contents. This impacted three components: console.anthropic.com, api.anthropic.com, and claude.ai. The incident began at Apr 10, 2025 5:48 PM UTC and was resolved by Apr 10, 2025 7:28 PM UTC, lasting about 1 hour and 40 minutes.

The Claude status page reported the issue and that engineers were applying a fix. The incident saw 2 official updates during its course and services returned to normal after the fix was deployed.

Anthropic Claude 3.7 Sonnet experienced a brief degraded service affecting the web console, API access, and Claude's interface. The degradation began at Apr 10, 2025 2:34 AM UTC and was resolved by Apr 10, 2025 2:58 AM UTC, lasting about 24 minutes. Users attempting to use Claude 3.7 Sonnet may have encountered elevated errors across console.anthropic.com, api.anthropic.com, and claude.ai.

Anthropic experienced elevated errors on requests to Claude 3.5 Sonnet models (June and October versions), initially impacting the api.anthropic.com endpoint. The issue later expanded to Claude 3.7 Sonnet as updates indicated ongoing mitigation.

The incident began at 4:28 PM UTC and was resolved by 8:20 PM UTC. The provider issued 3 status updates during the incident as teams investigated and mitigated the problem, with affected components including api.anthropic.com, console.anthropic.com, and claude.ai.

Anthropic Claude experienced a brief degraded performance impacting claude.ai, console.anthropic.com, and api.anthropic.com. The incident began at Apr 8, 2025 2:28 PM UTC and lasted approximately 1 hour 35 minutes, with services restored by Apr 8, 2025 4:03 PM UTC. A fix was implemented and the team began monitoring results, with two status updates issued as the incident progressed.

Anthropic Claude services, including claude.ai, api.anthropic.com, and console.anthropic.com, experienced elevated errors when handling long-context requests to Claude 3.7 Sonnet, causing intermittent failures or timeouts for users with long prompts or workflows. The outage lasted about 10 minutes, from 10:38 PM UTC to 10:48 PM UTC. The issue was identified and a fix was being implemented during the incident.

The provider published two status updates during the incident: the first noting elevated errors and that a fix was being implemented, and the second confirming restoration of service.