Increased processing times and errors for Universal-3.5 Pro on the EU async endpoint
major- Started
- Oct 5, 6:21 AM EDT
- Resolved
- Oct 5, 7:39 AM EDT
- Duration
- 1h 18m
AssemblyAI's updates (4)
- investigating · Oct 5, 6:21 AM EDT
We have identified an issue affecting asynchronous transcription requests using Universal-3.5 Pro on our EU endpoint. Most of these requests are currently significantly delayed, and some have failed. Our team is working to restore normal processing. Our US endpoint and real-time services are not affected. We will continue to provide updates as more information becomes available.
- identified · Oct 5, 6:38 AM EDT
We have identified the issue causing longer-than-usual processing times and some failed requests for asynchronous transcription using our Universal-3 Pro and Universal-3.5 Pro models on the EU endpoint. We are working to resolve this as quickly as possible and will provide further updates as soon as we can.
- monitoring · Oct 5, 6:43 AM EDT
We have continued to work to resolve this issue and are now seeing processing times return to the normal ranges, but we are continuing to monitor to ensure good performance.
- resolved · Oct 5, 7:39 AM EDT
This incident has been resolved. Between approximately 08:20 and 10:45 UTC, asynchronous transcription requests using Universal-3 Pro and Universal-3.5 Pro on our EU endpoint took longer than usual to complete, and some requests failed. Processing times have returned to normal ranges. Any requests that failed during this window can be safely resubmitted.
What our checks recorded
We checked AssemblyAI 15 times from AssemblyAI's start time to its resolution, once every five minutes. Every one of them reached AssemblyAI's site, so this incident did not show up as a failed page load. Our check loads the public site; it cannot see an API error or a broken feature. Median response time was 104 ms, against 108 ms over the 24 hours before. AssemblyAI posted the incident at Oct 5, 6:21 AM EDT.
Updates are AssemblyAI's own words from its status page (status: resolved); we never add an inferred cause. The vendor's incident page ↗