Is Anthropic (Claude) down?
Right now
Yes. Anthropic (Claude) has a partial outage.
Elevated errors on platform.claude.com.
Read from status.claude.com · every 60s
For agents
fallback
Use a fallback, next attempt in 300s.
GET upbutler.com/status/anthropic/status.agent.json
MCP: upstream_status {"service":"anthropic"}
The Claude API, Claude Code and claude.ai. Overload shows as HTTP 529 before the status page changes, so an independent probe matters here.
What Anthropic (Claude) reports
Partial outage· 1 open incident
Taken as is from Anthropic (Claude)'s own status page API. Incident titles and component names below are theirs.
Open status.claude.comUpButler's independent probe
API edge is answeringHTTP 401 · 137 ms
One request without credentials to api.anthropic.com every two minutes. An answer (the expected "not authenticated") shows the edge, TLS and routing work. It does not prove that authenticated calls succeed.
Checked · up since 2026-10-10 01:23 UTC
Open at Anthropic (Claude)
- Elevated errors on platform.claude.commonitoring · impact major
Since · Claude Console (platform.claude.com)
We've mitigated the errors loading usage data on platform.claude.com and in the Admin API usage report. While we work on a full fix, the most recent ~30 minutes of usage may take longer to appear on the platform.claude.com. API requests are not affected, and we will provide additional updates here.
Text written by Anthropic (Claude), shown unchanged.
Anthropic (Claude) components
1 of 6 not operational
- claude.aiOperational
- Claude Console (platform.claude.com)Degraded
- Claude API (api.anthropic.com)Operational
- Claude CodeOperational
- Claude CoworkOperational
- Claude for GovernmentOperational
Anthropic (Claude) incidents, last 90 days
6 finished incidents seen since UpButler started tracking on 2026-10-10, 4 of them rated major or critical by Anthropic (Claude). A typical one lasted 1 h 13 min; the longest 6 h 16 min.
| Started (UTC) | Incident | Impact | Lasted |
|---|---|---|---|
| Issue with spend limits | none | under a minute | |
| Elevated errors loading usage data on platform.claude.com | major | 1 h 13 min | |
| Elevated errors for Claude Opus 5.5 | major | 19 min | |
| Elevated errors for Claude Mythos 5.1 and Claude Fable 5.1 | major | 16 min | |
| Delayed credits on the Claude Platform | minor | 6 h 16 min | |
| Elevated errors on claude.ai, Claude Code, Claude Cowork and the Claude API | major | 2 h 6 min |
Tell your outage from Anthropic (Claude)'s
Declare that your project depends on Anthropic (Claude). When one of your monitors fails while Anthropic (Claude) has an incident that began around the same time, UpButler tags your incident "likely upstream" with their incident linked, says so in the alert and the AI report, and tells your on-call agent not to change code.
npx upbutler init finds the dependency from your env var names and packages. The catalog, dependencies and upstream-aware incidents are on every plan, including Free.
# upbutler.yaml
dependencies:
- anthropiccurl -s https://upbutler.com/status/anthropic/status.agent.json{
"schema": "upbutler.upstream-status/v1",
"service": "anthropic",
"verdict": "fallback",
"status": "partial_outage",
"reason": "Elevated errors on platform.claude.com",
"retryAfterSec": 300,
"incident": {
"title": "Elevated errors on platform.claude.com",
"url": "https://status.claude.com/incidents/978mjkgw6mh7"
},
"probe": {
"state": "up",
"statusCode": 401
}
}Questions
- Is Anthropic (Claude) down right now?
- Yes. As of 2026-10-10 05:12 UTC, Anthropic (Claude) has a partial outage, with an open incident: "Elevated errors on platform.claude.com". This page reads the status again every 60 seconds.
- Where does this status come from?
- From Anthropic (Claude)'s own status page API at https://status.claude.com. UpButler reads it once for everyone (not once per visitor) every 60 seconds with conditional requests, and shows it unchanged: incident titles and component names are Anthropic (Claude)'s words.
- What does UpButler's own Anthropic (Claude) check prove?
- Every two minutes UpButler sends one request without credentials to the Anthropic (Claude) API edge (api.anthropic.com). Any answer from the edge, usually HTTP 401, shows that DNS, TLS and routing work. It does not prove that authenticated requests succeed, so it can say "reachable" while Anthropic (Claude) reports degraded service. It is useful the other way round: when the probe fails before their status page changes.
- How can my agent or app check Anthropic (Claude) status?
- Fetch https://upbutler.com/status/anthropic/status.agent.json: a small JSON document with a verdict (proceed, retry, fallback or pause) and retryAfterSec. No key is needed. The same answer is an MCP tool, upstream_status {"service":"anthropic"}, on https://upbutler.com/mcp/public.
- How do I know whether an outage is mine or Anthropic (Claude)'s?
- Add "anthropic" to the dependencies of your UpButler workspace. When one of your monitors fails while Anthropic (Claude) has an incident that began around the same time, your incident is tagged "likely upstream" with their incident linked, and your on-call agent is told not to change code.
More in ai models and tooling
UpButler is not affiliated with Anthropic (Claude). Status and incident text belong to Anthropic (Claude) and are read from their public status source; the probe result is UpButler's own measurement.