# AI API outages on September 21, 2026 (UTC)

Measured every 5 minutes from 4 regions. An outage is attributed to a provider only when it failed from 2+ regions at once and no other provider failed in the same minutes.

## Was the Gemini API (Google) down on September 21, 2026?

Yes, measured. 3 outage(s) attributed to Google: requests to gemini-flash-lite-latest failed from up to 4 of 4 regions at once while no other provider was failing, first at 01:15 UTC, at least 6 minutes observed in total. At least once Google's own host answered with an HTTP 5xx.

| UTC | What failed | Regions | Duration (lower bound) | Failed probes | Verdict |
| --- | --- | --- | --- | --- | --- |
| 01:15-01:18 | requests to gemini-flash-lite-latest | ap-tokyo eu-hetzner sa-east us-central | >=3 min | 4 | provider_outage |
| 14:15-14:18 | requests to gemini-flash-lite-latest | eu-hetzner sa-east | >=3 min | 2 | unattributed |
| 14:35-14:35 | requests to gemini-flash-lite-latest | ap-tokyo sa-east us-central | <5 min | 3 | unattributed |
| 14:55-14:58 | requests to gemini-flash-lite-latest | ap-tokyo eu-hetzner sa-east us-central | >=3 min | 4 | provider_outage |
| 15:15-15:18 | requests to gemini-flash-lite-latest | ap-tokyo eu-hetzner sa-east us-central | >=3 min | 4 | unattributed |
| 19:15-19:15 | requests to gemini-flash-lite-latest | sa-east us-central | <5 min | 2 | provider_outage |

Live status now: https://llmlatency.dev/provider/google

## Was the featherless API down on September 21, 2026?

Yes, measured. 2 outage(s) attributed to featherless: host probes failed from up to 2 of 4 regions at once while no other provider was failing, first at 01:53 UTC, at least 4 minutes observed in total.

| UTC | What failed | Regions | Duration (lower bound) | Failed probes | Verdict |
| --- | --- | --- | --- | --- | --- |
| 01:53-01:55 | host probes | eu-hetzner us-central | >=2 min | 2 | provider_outage |
| 03:20-03:20 | host probes | ap-tokyo us-central | <5 min | 2 | unattributed |
| 03:40-04:10 | host probes | eu-hetzner sa-east us-central | >=30 min | 7 | unattributed |
| 04:50-04:55 | host probes | ap-tokyo us-central | >=5 min | 2 | unattributed |
| 05:23-06:00 | host probes | ap-tokyo eu-hetzner sa-east us-central | >=37 min | 8 | unattributed |
| 09:00-09:15 | host probes | ap-tokyo sa-east | >=15 min | 4 | unattributed |
| 10:03-10:05 | host probes | ap-tokyo eu-hetzner | >=2 min | 2 | provider_outage |
| 10:25-10:38 | host probes | eu-hetzner us-central | >=13 min | 2 | unattributed |
| 12:05-12:20 | host probes | ap-tokyo sa-east | >=15 min | 2 | unattributed |
| 14:13-14:25 | host probes | eu-hetzner us-central | >=12 min | 2 | unattributed |
| 16:25-16:45 | host probes | ap-tokyo sa-east us-central | >=20 min | 3 | unattributed |
| 18:05-19:55 | host probes | ap-tokyo eu-hetzner sa-east us-central | >=110 min | 28 | unattributed |

Live status now: https://llmlatency.dev/provider/featherless

## Was the baseten API down on September 21, 2026?

Not by our measurements. No failure of baseten was seen from two or more regions at once. baseten's own status page opened 2 entries that day (self-reported, listed below).
- self-reported 13:31 UTC: Metrics degradation in UI and API. **Inference not affected** (minor)
- self-reported 18:34 UTC: Elevated inference 500s for a subset of models deployed in GCP central1 (major)

Live status now: https://llmlatency.dev/provider/baseten

## Was the iflytek API down on September 21, 2026?

Not attributably. 1 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 (lower bound) | Failed probes | Verdict |
| --- | --- | --- | --- | --- | --- |
| 14:03-14:05 | host probes | eu-hetzner sa-east | >=2 min | 2 | unattributed |

Live status now: https://llmlatency.dev/provider/iflytek

## Was the minimax API down on September 21, 2026?

Not attributably. 1 multi-region failure(s) of minimax coincided with failures of other providers, so the cause sat above minimax and is not blamed on it.

| UTC | What failed | Regions | Duration (lower bound) | Failed probes | Verdict |
| --- | --- | --- | --- | --- | --- |
| 13:10-13:15 | host probes | sa-east us-central | >=5 min | 3 | unattributed |

Live status now: https://llmlatency.dev/provider/minimax

## Was the yi-01ai API down on September 21, 2026?

Not attributably. 2 multi-region failure(s) of yi-01ai coincided with failures of other providers, so the cause sat above yi-01ai and is not blamed on it.

| UTC | What failed | Regions | Duration (lower bound) | Failed probes | Verdict |
| --- | --- | --- | --- | --- | --- |
| 05:00-05:25 | host probes | ap-tokyo us-central | >=25 min | 4 | unattributed |
| 14:05-14:25 | host probes | ap-tokyo us-central | >=20 min | 3 | unattributed |

Live status now: https://llmlatency.dev/provider/yi-01ai

Source: https://llmlatency.dev/incidents/2026-09-21 · CC BY 4.0