Anthropic
All Systems Operational
- claude.aiOperational
- Claude Console (platform.claude.com)Operational
- Claude API (api.anthropic.com)Operational
- Claude CodeOperational
- Claude CoworkOperational
- Claude for GovernmentOperational
Live status and recent incidents for the AI stack most product teams build on, in one place. Refreshed every 5 minutes from each vendor’s official status API or feed. Subscribe to the aggregated RSS feed to get every new incident in one stream.
Rollup health and per-component status for every tracked vendor. Click through to the source status page for the official incident thread.
No components reported.
View official status pageNo components reported.
View official status pageActive incidents first, then resolved incidents from the last 30 days across every tracked vendor.
From 18:12–18:55, we saw elevated errors on requests to Claude Opus 4.8. This issue has been resolved.
This incident has been resolved.
All impacted services have now fully recovered.
This incident has been resolved.
All impacted services have now fully recovered.
All impacted services have now fully recovered.
Incorrectly suspended users have been restored access and received an email about their account and subscription information, if relevant. Impacted services have recovered.
All impacted services have now fully recovered.
All impacted services have now fully recovered.
Service Restored — Wed, 03 Jun 2026 10:46:42 GMT
Service Restored — Wed, 03 Jun 2026 11:34:55 GMT
This incident has been resolved.
This incident has been resolved.
This incident has been resolved.
This incident has been resolved.
This issue has been resolved.
This incident has been resolved.
This incident has been resolved.
This incident has been resolved.
This incident has been resolved.
An AI stack status tracker is a single dashboard that pulls the live operational status and recent incidents for every major AI provider in one view. This page aggregates the official status feeds of OpenAI, Anthropic, Google AI Studio, xAI, and Cursor, refreshes every five minutes, and surfaces a correlation banner when two or more vendors are degraded at the same time, the most common shape of a real AI outage.
Each provider card shows the current overall status (operational, degraded, partial outage, or major outage), the timestamp of the last refresh, and the most recent incident with its severity and resolution state. A timeline below the cards plots every incident from the last 30 days across all providers on a shared X-axis, so you can spot a cluster instantly when one cloud takes down multiple services.
Why bother? Because most AI app outages today are not your code. They are an upstream API hiccup. Having a single tab to glance at saves you ten minutes of "is it just me?" debugging every time the latency goes up. The same data is also exposed as an aggregated RSS feed so you can wire it into Slack, Discord, or a monitoring tool.
When the API Is Down
Frontier APIs go down. Local hardware does not. Find out which open-source AI models can run on the hardware you already own and what to switch to when OpenAI or Anthropic has an incident.
Per-token prices for the same providers tracked here — refreshed hourly.
Every open and closed model these providers ship, ranked by benchmark score.
When an API is down, this tool tells you whether local hardware would pay off.
Cheapest cloud GPU rentals across RunPod and Vast.ai for an emergency local fallback.
Every 5 minutes. The page calls each vendor’s official status API or RSS feed, normalizes the response, and caches the result for 5 minutes so a high-traffic moment never overwhelms the upstream services. The Last Refreshed label in the hero shows the exact timestamp of the latest pull.
Anthropic, OpenAI, and Cursor all run public Statuspage v2 APIs at status.anthropic.com, status.openai.com, and status.cursor.com. Google AI Studio runs on Google Cloud, so we pull AI-related services from status.cloud.google.com. xAI runs on Instatus and blocks JSON endpoints, so we read its public RSS feed at status.x.ai/feed.xml. We do not scrape, alter, or interpret the data beyond normalizing severity labels into a shared scale.
Cursor sits between most engineering teams and the underlying AI providers, so a Cursor outage takes your editor down even when the model APIs are healthy. We track it on the same page so you can answer "is it me, my IDE, or the model" in one glance.
Operational means no known issues. Minor means at least one component has a small degradation. Major means a component is partially unavailable. Critical means a major outage is in progress. The labels mirror Statuspage’s own severity scheme so they match what each vendor publishes on their own status page.
Use the Subscribe via RSS button in the hero. The feed includes active incidents and any resolved incidents from the last 7 days across every tracked vendor, in one stream, so you can wire it into Slack, Feedly, an email automation, or any RSS reader.
When two or more vendors are simultaneously reporting non-operational status and the earliest active incident started in the last 6 hours, we surface a banner at the top of the page. It is a fast tell that the issue might be cross-vendor rather than specific to one product, which often points at a shared upstream like cloud infrastructure or DNS.
We track vendors and their components, not individual model endpoints. A provider-level outage in OpenAI’s API or Anthropic’s API affects every model behind it. If you need model-specific reliability, the Made By Agents reference model database is the better resource and we plan to add per-model uptime as a follow-up.
Embed support is on the roadmap and will reuse the existing embed pattern from the AI Hardware Calculator and ROI Calculator. For now, link to the page directly or subscribe to the RSS feed and surface the latest incidents in your own UI.