Claude API disruption on 4 October 2026

Claude API from Amsterdam, Mumbai, San Francisco, Singapore, Tokyo, Montreal: degraded checks; 2026-10-04 13:30 UTC to 2026-10-04 15:30 UTC: peak first visible token 8.32 s vs a usual 1.34 s.

What our checks saw: Degraded
Cities affected: Amsterdam, Mumbai, San Francisco, Singapore, Tokyo, Montreal
When: 4 Oct, 13:30 UTC to 4 Oct, 15:30 UTC
How long: 2 hours
Slowest check: 8.32 s, usually 1.34 s
Anthropic's status page: Open it

Where our checks saw it

Each glowing city had slow or failing checks during the incident, shown at its worst. We test from 6 cities, so other places may have been hit too.

Our checks, city by city

Each point is one check every 30 minutes, from three hours before the incident to three hours after. Coloured points are the checks we counted as slower or failing. The shaded area shows how far above the usual time they were.

Amsterdam

Each check (UTC)Usual (1.31 s)Much slower, or failingIncident

Every check as a table

Time (UTC)	Check	Reading
10:30	1.37 s	Normal
11:00	1.19 s	Normal
11:30	1.10 s	Normal
12:00	1.23 s	Normal
12:30	1.42 s	Normal
13:00	1.40 s	Normal
13:30	14.17 s	Much slower, or failing
14:00	1.46 s	Normal
14:30	1.39 s	Normal
15:00	1.44 s	Normal
15:30	1.23 s	Normal
16:00	1.34 s	Normal
16:30	1.17 s	Normal
17:00	1.17 s	Normal
17:30	1.20 s	Normal
18:00	1.28 s	Normal
18:30	1.32 s	Normal

San Francisco

Each check (UTC)Usual (1.16 s)Much slower, or failingIncident

Every check as a table

Time (UTC)	Check	Reading
10:30	1.02 s	Normal
11:00	1.09 s	Normal
11:30	1.04 s	Normal
12:00	1.08 s	Normal
12:30	1.06 s	Normal
13:00	1.14 s	Normal
13:30	9.35 s	Much slower, or failing
14:00	1.47 s	Normal
14:30	1.23 s	Normal
15:00	1.29 s	Normal
15:30	1.16 s	Normal
16:00	1.47 s	Normal
16:30	1.08 s	Normal
17:00	1.09 s	Normal
17:30	1.20 s	Normal
18:00	1.09 s	Normal
18:30	1.07 s	Normal

Montreal

Each check (UTC)Usual (1.16 s)Much slower, or failingIncident

Every check as a table

Time (UTC)	Check	Reading
10:30	993 ms	Normal
11:00	1.10 s	Normal
11:30	1.04 s	Normal
12:00	1.22 s	Normal
12:30	1.08 s	Normal
13:00	1.04 s	Normal
13:30	14.49 s	Much slower, or failing
14:00	1.39 s	Normal
14:30	1.19 s	Normal
15:00	1.18 s	Normal
15:30	1.12 s	Normal
16:00	1.34 s	Normal
16:30	1.05 s	Normal
17:00	1.13 s	Normal
17:30	1.08 s	Normal
18:00	1.07 s	Normal
18:30	1.22 s	Normal

Singapore

Each check (UTC)Usual (1.33 s)Much slower, or failingIncident

Every check as a table

Time (UTC)	Check	Reading
10:30	1.23 s	Normal
11:00	1.26 s	Normal
11:30	1.19 s	Normal
12:00	1.23 s	Our check could not be verified
12:30	1.27 s	Normal
13:00	1.39 s	Normal
13:30	14.30 s	Much slower, or failing
14:00	1.52 s	Normal
14:30	1.43 s	Normal
15:00	1.45 s	Normal
15:30	1.29 s	Normal
16:00	1.56 s	Normal
16:30	1.32 s	Normal
17:00	1.23 s	Normal
17:30	1.27 s	Normal
18:00	1.27 s	Normal
18:30	1.25 s	Normal

Tokyo

Each check (UTC)Usual (1.21 s)Much slower, or failingIncident

Every check as a table

Time (UTC)	Check	Reading
10:30	1.14 s	Normal
11:00	1.15 s	Normal
11:30	1.17 s	Normal
12:00	1.17 s	Normal
12:30	1.17 s	Normal
13:00	1.27 s	Normal
13:30	9.31 s	Much slower, or failing
14:00	1.73 s	Normal
14:30	1.39 s	Normal
15:00	1.24 s	Normal
15:30	1.31 s	Normal
16:00	1.36 s	Normal
16:30	1.14 s	Normal
17:00	1.12 s	Normal
17:30	1.17 s	Normal
18:00	1.17 s	Normal
18:30	1.32 s	Normal

Mumbai

Each check (UTC)Usual (1.34 s)Slower than usualMuch slower, or failingIncident

Every check as a table

Time (UTC)	Check	Reading
10:30	1.16 s	Normal
11:00	1.35 s	Normal
11:30	1.24 s	Normal
12:00	1.26 s	Normal
12:30	1.39 s	Normal
13:00	1.82 s	Normal
13:30	14.35 s	Much slower, or failing
14:00	2.29 s	Slower than usual
14:30	1.54 s	Normal
15:00	1.31 s	Normal
15:30	1.36 s	Normal
16:00	1.47 s	Normal
16:30	1.26 s	Normal
17:00	1.24 s	Normal
17:30	1.28 s	Normal
18:00	1.34 s	Normal
18:30	1.42 s	Normal

How fast is your API around the world?

Run a free speed test from multiple cities and find out where your users are waiting. No setup, no account required.

Test my API

No account required · Takes about 30 seconds.