AI API outages on September 27, 2026
UTC day · measured every 5 minutes from 4 regions · method
On September 27, 2026 (UTC) our probes attributed 4 outage(s) to 3 provider(s): Gemini API (Google): 2; aleph-alpha API: 1; doubao API: 1.
This page is history. Is it down right now?
Live status, updated every 30 minutes: Gemini API (Google), aleph-alpha API, doubao API, baichuan API, stepfun API · all AI APIs, last 72 hours
Was the Gemini API (Google) down on September 27, 2026?
Yes, measured. 2 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 05:05 UTC, each shorter than one 5-minute probe cycle.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 05:05–05:05 | requests to gemini-flash-lite-latest | 2 of 4: South America (São Paulo), US (Central) | <5 min | 2 | provider outage | timeout |
| 12:36–13:06 | requests to gemini-flash-lite-latest | 3 of 4: Asia (Tokyo), Europe (Germany), US (Central) | ≥30 min | 4 | shared with 4 other provider(s) — not attributed | timeout |
| 13:25–13:25 | requests to gemini-flash-lite-latest | 2 of 4: South America (São Paulo), US (Central) | <5 min | 2 | shared with 4 other provider(s) — not attributed | timeout |
| 14:05–14:17 | requests to gemini-flash-lite-latest | 2 of 4: Europe (Germany), US (Central) | ≥13 min | 2 | shared with 4 other provider(s) — not attributed | timeout |
| 14:45–15:05 | requests to gemini-flash-lite-latest | 2 of 4: Europe (Germany), South America (São Paulo) | ≥20 min | 3 | shared with 2 other provider(s) — not attributed | timeout |
| 15:25–15:25 | requests to gemini-flash-lite-latest | 2 of 4: Asia (Tokyo), US (Central) | <5 min | 2 | provider outage | timeout |
| 16:05–16:17 | requests to gemini-flash-lite-latest | 3 of 4: Asia (Tokyo), Europe (Germany), US (Central) | ≥13 min | 3 | shared with 1 other provider(s) — not attributed | timeout |
Current Gemini API (Google) status
Was the aleph-alpha API down on September 27, 2026?
Yes, measured. 1 outage(s) attributed to aleph-alpha: host probes failed from up to 2 of 4 regions at once while no other provider was failing, first at 19:15 UTC, each shorter than one 5-minute probe cycle.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 19:15–19:15 | host probes | 2 of 4: Asia (Tokyo), South America (São Paulo) | <5 min | 2 | provider outage | timeout |
Current aleph-alpha API status
Was the doubao API down on September 27, 2026?
Yes, measured. 1 outage(s) attributed to doubao: host probes failed from up to 2 of 4 regions at once while no other provider was failing, first at 04:30 UTC, each shorter than one 5-minute probe cycle.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 04:30–04:31 | host probes | 2 of 4: Europe (Germany), US (Central) | <5 min | 2 | provider outage | timeout |
| 13:23–13:25 | host probes | 2 of 4: Europe (Germany), South America (São Paulo) | ≥2 min | 2 | shared with 4 other provider(s) — not attributed | timeout |
| 14:45–14:45 | host probes | 2 of 4: South America (São Paulo), US (Central) | <5 min | 2 | shared with 1 other provider(s) — not attributed | timeout |
Was the baichuan API down on September 27, 2026?
Not attributably. 1 multi-region failure(s) of baichuan coincided with failures of other providers, so the cause sat above baichuan and is not blamed on it.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 13:06–13:06 | host probes | 2 of 4: South America (São Paulo), US (Central) | <5 min | 2 | shared with 3 other provider(s) — not attributed | timeout |
Was the stepfun API down on September 27, 2026?
Not attributably. 1 multi-region failure(s) of stepfun coincided with failures of other providers, so the cause sat above stepfun and is not blamed on it.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 03:50–04:01 | host probes | 4 of 4: Asia (Tokyo), Europe (Germany), South America (São Paulo), US (Central) | ≥11 min | 5 | shared with 1 other provider(s) — not attributed | [Errno 101] Network is unreachable; timeout |
55 single-region failure(s) were also recorded that day (iflytek 12, aleph-alpha 12, baichuan 11, google 5, stepfun 4, sensenova 3, hunyuan 3, doubao 2, qwen 1, openai 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: Sep 26 · Last 72 hours · Next: Sep 28
Machine-readable: incidents.json · Licence: CC BY 4.0