Add named vectors scenario to embedding migration tutorial (#2348)

* Initial commit

* Edits

* Update FAQ

* Edits

* Mention required versions
This commit is contained in:
Abdon Pijpelink
2026-05-26 09:38:24 +02:00
committed by GitHub
parent b889e19117
commit 04e3e0bb1d
52 changed files with 1967 additions and 35 deletions
+1 -1
View File
@@ -101,7 +101,7 @@ The same data is available [through Qdrant's API](/documentation/ops-monitoring/
When a collection's vector schema needed to change, for example, when adding support for a new embedding model or retiring an old one, there was no way to update it in place. The only option was to recreate the collection from scratch and re-ingest all points.
Version 1.18 makes it possible to [add and remove named vectors](/documentation/manage-data/collections/#update-vectors) to an existing collection's schema without having to recreate it. This makes embedding model migration considerably easier. You can now add a new vector field, populate it in the background, and remove the old one when you're ready.
Version 1.18 makes it possible to [add and remove named vectors](/documentation/manage-data/collections/#update-vector-schema) to an existing collection's schema without having to recreate it. This makes embedding model migration considerably easier. You can now add a new vector field, populate it in the background, and remove the old one when you're ready.
## Audit Logging Improvements