App status explained · independent evidence

Is Claude down right now?

Something feels slow, broken or unavailable in Claude? Check the source and freshness before drawing a conclusion.

What the evidence says

Monitoring not activated. No current independent evidence.

Official Claude app: Official status unknown
Official Anthropic API: Official status unknown
Independent Anthropic API: Insufficient evidence

API checks do not measure the consumer app.

Monitoring not activated. No verified current independent snapshot is available.

Official Claude / claude.ai status

Official status unknown

No verified current component evidence.

Source checked: Unavailable · unavailable

Official Anthropic / Claude API status

Official status unknown

No verified current component evidence.

Source checked: Unavailable · unavailable

Independent Anthropic / Claude API checks

Insufficient evidence

Latency comparison: insufficient evidence

Latest check slot: Unavailable
0 / 3 fresh regions · current regional evidence unavailable

API checks do not measure the consumer app. Official reports and independent observations may disagree. Missing evidence does not mean an outage.

Independent checks: every 5 minutes when activated; unavailable after a missed expected check plus 60 seconds. Official evidence expires after 90 seconds. Each source ages separately.

Open official provider status · Provider update: Unavailable

What should I understand next?

If Claude is not working, check whether an official claude.ai incident matches your problem. An API failure or slowdown alone does not establish a consumer-app outage.

Anthropic API evidence for developers · How we measure and handle missing evidence

Regional API evidence

Anthropic / Claude

Insufficient evidence

Official API: unknown
Overall provider: unknown · fetched No verified check

Official incident coverage: unknown · Official source

Consumer components: unknown; separate from API checks.

Independent: No verified check
0 / 3 fresh regions · 0 responding

SeoulUnavailableLast checked No verified check · baseline insufficient
N. VirginiaUnavailableLast checked No verified check · baseline insufficient
OregonUnavailableLast checked No verified check · baseline insufficient

claude-haiku-4-5-20251001 · health-v1 · first readable text. Representative API model only; consumer apps and provider-wide health are not measured

Source: AI Fast Now independent API checks. Expected every 5 min; unavailable after a missed expected check (60 sec dispatch grace).

Read the failure category and model before interpreting a result. Failed requests have no invented zero latency. Historical or stale checks do not establish present availability.

What these sources cover

The official claude.ai component describes the consumer application. Claude API is a separate component. Our representative Anthropic API check does not test your claude.ai conversation or account.

Official consumer components, official API components and independent API checks are three separate sources. An overall provider indicator is not a substitute for a matching component. No incident reported is limited to the supported official components and fresh supplied incident coverage; it is not proof that every user can connect.

Why can the service feel slow without an incident?

A long conversation, a different model, account limits, a local network path or app routing may affect your experience. These are possible explanations, not diagnoses. We only label elevated API latency after repeated comparable checks exceed a distributed baseline. Without enough baseline evidence, speed remains inconclusive.

One successful request does not rule out intermittent failures. One failed request does not establish a consumer-app outage. Missing regions and expired observations remain visible.

Model, region and measurement details

Current-health representative model: claude-haiku-4-5-20251001. The health-v1 profile measures time to first readable API text. Benchmark speed requests are a separate historical experiment.

Anthropic API provider details and model catalog · Seoul · N. Virginia · Oregon

Regions identify the origins of our probes, not provider server locations or the route taken by your browser. Read timestamps as UTC; a recently published page does not refresh an old measurement.

Performance history

Reports describe completed API measurement periods, not current consumer-app status.

Daily: no complete verified report yet
Weekly: no complete verified report yet
Monthly: no complete verified report yet

Archive: verified historical evidence

Earlier bounded observation (not a complete Daily, Weekly or Monthly report): AI Performance Observation Report: 2026-10-02–2026-10-03

Calendar reports use America/New_York local dates, including daylight-saving changes. Publication requires a closed period, complete expected slots, immutable source verification, reconciled accounting and explicit publication permission. Partial periods are not completed reports.

Measurement methodology · Editorial policy