Is Claude API down? › Outage history
Claude API outage history
Claude API published 5 incidents on its official status page since Sep 1, 2026. 1 was a major or long incident with its own page; the biggest are listed first. Below them is every smaller incident, then our own 90 day check history.
Biggest Claude API outages
Ranked by the official impact level, then by how long they lasted. Each has its own page with the cause, the fix time and every official update.
- Oct 7, 2026: Elevated errors on platform.claude.com 6 days 5 hours · rated "Major"
Every smaller Claude API incident
Minor and short incidents from the Claude status page, newest first, each with a line from the official updates. Times are UTC.
October 2026
Oct 7, 2026, 01:20 UTC · Elevated errors loading usage data on platform.claude.com
1 hour 13 min · rated "Major"
"This incident has been resolved. From 17:00 PT on October 6 / 00:00 UTC on October 7 to 18:40 PT on October 6 / 01:40 UTC on October 7, many requests to load usage data on the platform.claude."
September 2026
Sep 29, 2026, 14:21 UTC · Elevated errors on claude.ai, Claude Code, Claude Cowork and the Claude API
2 hours 6 min · rated "Major"
"The issue affecting Claude.ai, the Claude desktop and mobile apps, Claude Code, Claude Cowork, and the Claude API has been resolved, and services have been operating normally since recovery."
Sep 10, 2026, 21:43 UTC · Elevated latency on the Claude API for some users
1 hour 41 min · rated "Minor"
"The issue affecting Claude services has been mitigated."
Sep 1, 2026, 17:05 UTC · Degraded performance on platform.claude.com and Claude for Microsoft Office 365
1 hour 2 min · rated "Minor"
"This issue has been resolved."
Claude API uptime, last 90 days
Tracked since Oct 7, 2026. Updated Oct 22, 2026, 12:35 UTC.
Every Claude API problem in the last 90 days
- Degraded
Elevated errors affecting Connectors
Oct 22, 2026, 03:52 UTC · lasted 4 hours 16 min · from Claude API's official status page
- Degraded
Degraded performance for Claude Fable 5
Oct 21, 2026, 19:52 UTC · lasted 48 minutes · from Claude API's official status page
- Degraded
Elevated errors on platform.claude.com
Oct 7, 2026, 21:22 UTC · lasted 140 hours 45 min · from Claude API's official status page
Day by day
| Day (UTC) | Uptime | Avg response | Worst |
|---|---|---|---|
| Oct 22, 2026 | 100% | 93 ms | Degraded |
| Oct 21, 2026 | 100% | 148 ms | Degraded |
| Oct 20, 2026 | 100% | 123 ms | Up |
| Oct 19, 2026 | 100% | 123 ms | Up |
| Oct 18, 2026 | 100% | 124 ms | Up |
| Oct 17, 2026 | 100% | 122 ms | Up |
| Oct 16, 2026 | 100% | 241 ms | Up |
| Oct 15, 2026 | 100% | 258 ms | Up |
| Oct 14, 2026 | 100% | 252 ms | Up |
| Oct 13, 2026 | 100% | 267 ms | Degraded |
| Oct 12, 2026 | 100% | 255 ms | Degraded |
| Oct 11, 2026 | 100% | 205 ms | Degraded |
| Oct 10, 2026 | 100% | 136 ms | Degraded |
| Oct 9, 2026 | 100% | 146 ms | Degraded |
| Oct 8, 2026 | 100% | 146 ms | Degraded |
| Oct 7, 2026 | 100% | 136 ms | Degraded |
Claude API outage questions
Was Claude API down today?
Yes. Claude API had a problem today (Oct 22, 2026, UTC): Elevated errors affecting Connectors (degraded, 4 hours 16 min).
Was Claude API down yesterday?
Yes. Claude API had a problem yesterday (Oct 21, 2026, UTC): Degraded performance for Claude Fable 5 (degraded, 48 minutes).
When was Claude API last down?
The latest problem was on Oct 22, 2026, 03:52 UTC and lasted 4 hours 16 min: Elevated errors affecting Connectors.
What was the biggest Claude API outage?
Of the incidents on record since Sep 1, 2026, the biggest was on Oct 7, 2026: Elevated errors on platform.claude.com. The Claude status page rated it "Major" and it lasted 6 days 5 hours.
How often does Claude API go down?
Claude API's status page posted 5 incidents since Sep 1, 2026, 3 of them in the last 30 days. Most incidents affect part of the service, not all of it. Uptime from our own checks over the last 90 days: 100%.
Times are UTC. Official incidents are the company's own posts, quoted from the Claude status page. Problem periods start when the official status page or our checks first show a problem and end when the service is next found working, so durations are accurate to within a few minutes. Recent outages across all services.