AI API outages on November 17, 2026
UTC day · measured every 5 minutes from 4 regions · method
On November 17, 2026 (UTC) our probes attributed 8 outage(s) to 3 provider(s): sensenova API: 6; Gemini API (Google): 1; baichuan API: 1. Providers’ own status pages opened 4 entries (Anthropic, Cerebras, OpenAI, replicate).
This page is history. Is it down right now?
Live status, updated every 30 minutes: sensenova API, Gemini API (Google), baichuan API, Claude API (Anthropic), Cerebras API, OpenAI API · all AI APIs, last 72 hours
Was the sensenova API down on November 17, 2026?
Yes, measured. 6 outage(s) attributed to sensenova: host probes failed from up to 4 of 4 regions at once while no other provider was failing, first at 00:15 UTC, at least 15 minutes observed in total.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 00:15–00:20 | host probes | 3 of 4: Asia (Tokyo), South America (Santiago), US (Central) | ≥5 min | 5 | provider outage | timeout |
| 02:20–02:20 | host probes | 2 of 4: Asia (Tokyo), US (Central) | <5 min | 2 | provider outage | timeout |
| 03:20–03:23 | host probes | 3 of 4: Asia (Tokyo), Europe (Germany), US (Central) | ≥3 min | 3 | provider outage | timeout |
| 20:20–20:23 | host probes | 2 of 4: Europe (Germany), South America (Santiago) | ≥3 min | 2 | provider outage | timeout |
| 21:23–21:25 | host probes | 4 of 4: Asia (Tokyo), Europe (Germany), South America (Santiago), US (Central) | ≥2 min | 4 | provider outage | timeout |
| 23:23–23:25 | host probes | 4 of 4: Asia (Tokyo), Europe (Germany), South America (Santiago), US (Central) | ≥2 min | 4 | provider outage | timeout |
Was the Gemini API (Google) down on November 17, 2026?
Yes, measured. 1 outage(s) attributed to Google: requests to gemini-flash-lite-latest failed from up to 2 of 4 regions at once while no other provider was failing, first at 09:45 UTC, at least 10 minutes observed in total.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 09:45–09:55 | requests to gemini-flash-lite-latest | 2 of 4: Asia (Tokyo), South America (Santiago) | ≥10 min | 2 | provider outage | timeout |
Current Gemini API (Google) status
Was the baichuan API down on November 17, 2026?
Yes, measured. 1 outage(s) attributed to baichuan: host probes failed from up to 2 of 4 regions at once while no other provider was failing, first at 11:43 UTC, at least 12 minutes observed in total.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 11:43–11:55 | host probes | 2 of 4: Asia (Tokyo), Europe (Germany) | ≥12 min | 2 | provider outage | timeout |
Was the Claude API (Anthropic) down on November 17, 2026?
Not by our measurements. No failure of Anthropic was seen from two or more regions at once. Anthropic’s own status page opened 1 entry that day (self-reported, listed below).
Self-reported on Anthropic’s status page, not verified by our probes:
- major Some users unable to sign in with Apple on claude.ai and the Claude iOS app · resolved Nov 17, 21:51 UTC
Current Claude API (Anthropic) status
Was the Cerebras API down on November 17, 2026?
Not by our measurements. No failure of Cerebras was seen from two or more regions at once. Cerebras’s own status page opened 1 entry that day (self-reported, listed below).
Self-reported on Cerebras’s status page, not verified by our probes:
- notice Partial Service Disruption · resolved Nov 17, 08:00 UTC
Was the OpenAI API down on November 17, 2026?
Not by our measurements. No failure of OpenAI was seen from two or more regions at once. OpenAI’s own status page opened 1 entry that day (self-reported, listed below).
Self-reported on OpenAI’s status page, not verified by our probes:
- minor Costs API data is delayed · resolved — UTC
Was the replicate API down on November 17, 2026?
Not attributably. 3 multi-region failure(s) of replicate coincided with failures of other providers, so the cause sat above replicate and is not blamed on it. replicate’s own status page opened 1 entry that day (self-reported, listed below).
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 13:13–13:20 | host probes | 2 of 4: Europe (Germany), South America (Santiago) | ≥7 min | 2 | shared with 1 other provider(s) — not attributed | http 502 |
| 14:10–14:10 | host probes | 2 of 4: South America (Santiago), US (Central) | <5 min | 2 | shared with 1 other provider(s) — not attributed | http 502 |
| 14:38–14:48 | host probes | 4 of 4: Asia (Tokyo), Europe (Germany), South America (Santiago), US (Central) | ≥10 min | 6 | shared with 1 other provider(s) — not attributed | http 502 |
Self-reported on replicate’s status page, not verified by our probes:
- minor Errors and stalled predictions on some models · resolved Nov 18, 03:23 UTC
Was the iflytek API down on November 17, 2026?
Not attributably. 2 multi-region failure(s) of iflytek coincided with failures of other providers, so the cause sat above iflytek and is not blamed on it.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 04:50–05:10 | host probes | 2 of 4: Asia (Tokyo), US (Central) | ≥20 min | 5 | shared with 2 other provider(s) — not attributed | timeout |
| 15:20–15:33 | host probes | 3 of 4: Asia (Tokyo), Europe (Germany), South America (Santiago) | ≥13 min | 3 | shared with 2 other provider(s) — not attributed | timeout |
38 single-region failure(s) were also recorded that day (doubao 9, google 8, iflytek 7, baichuan 6, replicate 3, sensenova 2, friendli 2, aleph-alpha 1). A failure seen from one region cannot be separated from a problem on that network path, so none of them is called an outage.
How is an outage told apart from a network problem?
Each provider is probed every five minutes from 4 independent regions. A failure is attributed to the provider only when it is seen from two or more regions at once and no other provider fails in the same minutes; a failure shared with other providers has a cause above all of them and is not blamed on anyone. Durations are lower bounds: we see the first and last failed probe, not the real start and end. Status-page entries are the provider’s own words, shown next to the measurement, not instead of it.
Previous: Nov 16 · Last 72 hours
Machine-readable: incidents.json · Licence: CC BY 4.0