Changes / Elastic Cloud / Increased error rates for microsoft-multilingual-e5-large

Elastic Cloud: Major incident: Increased error rates for microsoft-multilingual-e5-large

Summary

Elastic Cloud reported a major incident starting 2026-08-13 08:32 UTC, lasting 4 days 3 h, with search embedding failures for microsoft-multilingual-e5-large due to upstream provider issues. Resolved 2026-08-17 11:40 UTC, the model was retired from Elastic Inference Service; requests now return errors, and customers must create custom inference endpoints using their own API keys.

Machine-written from the diff below (Sol; type: incident outage). Check the diff before relying on it.

Major incidentLabel: outage (model (Jev), 99%)Changed Detected Source: stspg.io (opens in new tab)

What changed

AddedTitle#1
Before: not present
After: Increased error rates for microsoft-multilingual-e5-large
AddedImpact#2
Before: not present
After: major
AddedStatus#3
Before: not present
After: resolved
AddedStart time#4
Before: not present
After: 2026-08-13 08:32 UTC
AddedResolved time#5
Before: not present
After: 2026-08-17 11:40 UTC
AddedUpdateinvestigating#6
Before: not present
After: We are seeing elevated error rates from the upstream provider for the microsoft-multilingual-e5-large embedding model. Search embedding requests to this model may fail. We are investigating
AddedUpdateinvestigating#7
Before: not present
After: We're seeing some improvement in availability but are continuing to monitor the situation. We will follow up with more information in due course.
AddedUpdatemonitoring#8
Before: not present
After: Unfortunately, we can no longer support the E5 Multilingual Large model on the Elastic Inference Service due to continued upstream provider issues. Any customers requiring this model should set up custom inference endpoints using their own API keys.
AddedUpdateresolved#9
Before: not present
After: The microsoft-multilingual-e5-large model has been retired from the Elastic Inference Service. Requests to this model now return an error. Customers who wish to use this model with Elastic cloud must create a custom inference endpoint using their own API key. See our inference documentation for further details: https://www.elastic.co/docs/api/doc/elasticsearch/group/endpoint-inference

First version of this incident archived by Rewindex. View the live incident (opens in new tab).

History of this incident

When (UTC)ChangeKind
Major incident: Increased error rates for microsoft-multilingual-e5-large (this change)
Elastic Cloud reported a major incident starting 2026-08-13 08:32 UTC, lasting 4 days 3 h, with search embedding failures for microsoft-multilingual-e5-large due to upstream provider issues. Resolved 2026-08-17 11:40 UTC, the model was retired from Elastic Inference Service; requests now return errors, and customers must create custom inference endpoints using their own API keys.
Major incident

All Elastic Cloud changes → · Elastic Cloud profile · Markdown · JSON