AI API outages on August 3, 2026
UTC day · measured every 5 minutes from 4 regions · method
On August 3, 2026 (UTC) our probes attributed 14 outage(s) to 2 provider(s): friendli API: 13; aleph-alpha API: 1. Providers’ own status pages opened 2 entries (Anthropic).
This page is history. Is it down right now?
Live status, updated every 30 minutes: friendli API, aleph-alpha API, Claude API (Anthropic), sensenova API, stepfun API, 01.AI API · all AI APIs, last 72 hours
Was the friendli API down on August 3, 2026?
Yes, measured. 13 outage(s) attributed to friendli: host probes failed from up to 3 of 4 regions at once while no other provider was failing, first at 00:00 UTC, at least 147 minutes observed in total. At least once friendli’s own host answered with an HTTP 5xx.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 00:00–00:15 | host probes | 3 of 4: Asia (Tokyo), Europe (Germany), South America (São Paulo) | ≥15 min | 3 | provider outage | http 503 |
| 00:55–00:55 | host probes | 2 of 4: South America (São Paulo), US (Central) | <5 min | 2 | provider outage | http 503 |
| 01:20–01:25 | host probes | 3 of 4: Asia (Tokyo), Europe (Germany), US (Central) | ≥5 min | 3 | provider outage | http 503 |
| 03:15–03:50 | host probes | 4 of 4: Asia (Tokyo), Europe (Germany), South America (São Paulo), US (Central) | ≥35 min | 7 | shared with 4 other provider(s) — not attributed | http 503 |
| 04:20–04:30 | host probes | 3 of 4: Asia (Tokyo), South America (São Paulo), US (Central) | ≥10 min | 3 | provider outage | http 503 |
| 05:05–05:05 | host probes | 2 of 4: South America (São Paulo), US (Central) | <5 min | 2 | provider outage | http 503 |
| 07:55–08:20 | host probes | 3 of 4: Europe (Germany), South America (São Paulo), US (Central) | ≥26 min | 4 | shared with 2 other provider(s) — not attributed | http 503 |
| 08:50–09:00 | host probes | 2 of 4: Europe (Germany), South America (São Paulo) | ≥10 min | 2 | shared with 2 other provider(s) — not attributed | http 503 |
| 09:20–09:35 | host probes | 2 of 4: Asia (Tokyo), South America (São Paulo) | ≥15 min | 3 | shared with 3 other provider(s) — not attributed | http 503 |
| 09:55–10:20 | host probes | 2 of 4: Asia (Tokyo), US (Central) | ≥25 min | 5 | provider outage | http 503 |
| 11:40–11:45 | host probes | 2 of 4: Europe (Germany), US (Central) | ≥4 min | 2 | shared with 1 other provider(s) — not attributed | http 503 |
| 12:05–12:15 | host probes | 3 of 4: Europe (Germany), South America (São Paulo), US (Central) | ≥10 min | 3 | provider outage | http 503 |
| 13:15–13:30 | host probes | 2 of 4: South America (São Paulo), US (Central) | ≥15 min | 2 | provider outage | http 503 |
| 13:50–13:55 | host probes | 2 of 4: Europe (Germany), South America (São Paulo) | ≥5 min | 2 | shared with 3 other provider(s) — not attributed | http 503 |
| 14:20–14:30 | host probes | 2 of 4: Europe (Germany), South America (São Paulo) | ≥10 min | 2 | provider outage | http 503 |
| 16:45–17:00 | host probes | 2 of 4: Europe (Germany), South America (São Paulo) | ≥14 min | 2 | provider outage | http 503 |
| 17:44–18:00 | host probes | 2 of 4: Europe (Germany), US (Central) | ≥16 min | 3 | provider outage | http 503 |
| 18:55–19:20 | host probes | 4 of 4: Asia (Tokyo), Europe (Germany), South America (São Paulo), US (Central) | ≥25 min | 5 | shared with 2 other provider(s) — not attributed | http 503 |
| 22:50–23:05 | host probes | 2 of 4: South America (São Paulo), US (Central) | ≥15 min | 4 | provider outage | http 503 |
| 23:48–00:00 | host probes | 2 of 4: Asia (Tokyo), Europe (Germany) | ≥12 min | 3 | provider outage | http 503 |
Was the aleph-alpha API down on August 3, 2026?
Yes, measured. 1 outage(s) attributed to aleph-alpha: host probes failed from up to 3 of 4 regions at once while no other provider was failing, first at 16:10 UTC, each shorter than one 5-minute probe cycle. At least once aleph-alpha’s own host answered with an HTTP 5xx.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 16:10–16:10 | host probes | 3 of 4: Asia (Tokyo), South America (São Paulo), US (Central) | <5 min | 3 | provider outage | http 503 |
Current aleph-alpha API status
Was the Claude API (Anthropic) down on August 3, 2026?
Not by our measurements. No failure of Anthropic was seen from two or more regions at once. Anthropic’s own status page opened 2 entries that day (self-reported, listed below).
Self-reported on Anthropic’s status page, not verified by our probes:
- minor Error rates across multiple models · resolved Aug 3, 14:17 UTC
- minor Degraded performance on Claude Sonnet 5 · resolved Aug 3, 15:29 UTC
Current Claude API (Anthropic) status
Was the sensenova API down on August 3, 2026?
Not attributably. 2 multi-region failure(s) of sensenova coincided with failures of other providers, so the cause sat above sensenova and is not blamed on it.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 12:20–12:35 | host probes | 2 of 4: South America (São Paulo), US (Central) | ≥15 min | 3 | shared with 2 other provider(s) — not attributed | [Errno 104] Connection reset by peer; timeout |
| 14:05–14:05 | host probes | 2 of 4: Asia (Tokyo), South America (São Paulo) | <5 min | 2 | shared with 2 other provider(s) — not attributed | timeout |
Was the stepfun API down on August 3, 2026?
Not attributably. 2 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 |
|---|---|---|---|---|---|---|
| 11:20–11:25 | host probes | 2 of 4: South America (São Paulo), US (Central) | ≥5 min | 3 | shared with 2 other provider(s) — not attributed | [Errno 101] Network is unreachable; timeout |
| 14:05–14:05 | host probes | 2 of 4: South America (São Paulo), US (Central) | <5 min | 2 | shared with 2 other provider(s) — not attributed | timeout |
Was the 01.AI API down on August 3, 2026?
Not attributably. 3 multi-region failure(s) of 01.AI coincided with failures of other providers, so the cause sat above 01.AI and is not blamed on it.
| UTC | What failed | Regions | Duration | Failed probes | Verdict | Error |
|---|---|---|---|---|---|---|
| 11:35–11:40 | host probes | 2 of 4: Asia (Tokyo), South America (São Paulo) | ≥5 min | 2 | shared with 1 other provider(s) — not attributed | timeout |
| 12:20–13:05 | host probes | 2 of 4: Asia (Tokyo), South America (São Paulo) | ≥45 min | 5 | shared with 4 other provider(s) — not attributed | timeout |
| 16:25–16:35 | host probes | 2 of 4: Asia (Tokyo), South America (São Paulo) | ≥10 min | 2 | shared with 3 other provider(s) — not attributed | timeout |
77 single-region failure(s) were also recorded that day (yi-01ai 18, friendli 15, google 9, iflytek 7, sensenova 6, baichuan 6, doubao 5, stepfun 3, openai 2, featherless 2). 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: Aug 2 · Last 72 hours · Next: Aug 4
Machine-readable: incidents.json · Licence: CC BY 4.0