… visits today
ONLINE

Provider status

We ask every provider one small question every 10 minutes and record what comes back. This is that record, kept for 7 days.

All 6 providers answered the last checkchecked 10 minutes ago
answered slow did not answer not measured no check

Anthropic

Answering
100%
answered, 24h
98.63%
answered, 7d
1.2s
median reply
7 days agonow

144 of 144 checks measured in the last 24h

DeepSeek

Answering
100%
answered, 24h
99.42%
answered, 7d
0.9s
median reply
7 days agonow

144 of 144 checks measured in the last 24h

Z.ai (GLM)

Answering
100%
answered, 24h
99.9%
answered, 7d
1.8s
median reply
7 days agonow

144 of 144 checks measured in the last 24h

Google

Answering
100%
answered, 24h
97.85%
answered, 7d
0.6s
median reply
7 days agonow

144 of 144 checks measured in the last 24h

Moonshot (Kimi)

Answering
100%
answered, 24h
98.83%
answered, 7d
1.2s
median reply
7 days agonow

144 of 144 checks measured in the last 24h

OpenAI

Answering
100%
answered, 24h
98.24%
answered, 7d
1.1s
median reply
7 days agonow

144 of 144 checks measured in the last 24h

Outages we recorded

Checks we could not make

These are our failures, not the providers’. We publish them because leaving them out would turn our own expired keys, unpaid invoices, broken probes and server trouble into somebody else’s downtime — and because a gap in the record is a fact about the record. Each line says how long it lasted and when it last happened: anything not marked ongoing is over, and stays listed only until it falls out of the 7-day window. This list starts from 11 September 2026. Before that the provider accounts were unfunded and every provider returned a billing error; those checks are still recorded and still count against our uptime, but republishing a resolved funding gap for a week alongside live figures said something about today that was not true.

None in the last 7 days.

Slow periods

Two or more consecutive checks that took longer than 5 seconds. Measured from our machine, which is often running a benchmark sweep at the same time, so read these next to the median above rather than on their own.

How this is measured

The check
One short completion on our own key, from one machine in Germany. It is the same call for every provider, so the numbers are comparable.
What “answered” means
The provider returned usable content. It does not mean the model was correct, fast, or good — that is what the benchmarks are for.
When a check fails
We ask again fifteen seconds later and record the second answer, so one dropped connection is not published as ten minutes of outage. If the provider still cannot be reached, we test our own side before blaming theirs: whether our server stalled while the check was in flight, whether other providers failed to connect at the same moment, and whether two independent control endpoints answer. Any of those means the failure was ours — it is listed under checks we could not make and counts as neither uptime nor downtime.
What we do not claim
One probe, every 10 minutes, from one machine in Germany, on one account. A provider can be down for us and up for you, or the reverse. If your own traffic disagrees with this page, trust your own traffic.
When a provider is down
We stop benchmarking it rather than publishing a number we did not measure, and run the missed models as soon as it answers again. An outage leaves a gap in the charts — that gap is deliberate.

Generated 2026-10-08 19:59 UTC · refreshed every minute