$vectorSearch (aggregation)
$vectorSearch
performs a semantic search on data in your Atlas
cluster. If you store vector embeddings that are less than or equal to
4096 dimensions in width for any kind of data along with other data in
your collection on the Atlas cluster, you can seamlessly index the
vector data along with your other data. You can then use the
$vectorSearch
stage to pre-filter your data and perform
semantic search against the indexed fields.
Important
The $vectorSearch
aggregation pipeline stage is only available
for collections hosted on MongoDB Atlas cluster tiers
running MongoDB version 6.0.11 or later. To learn more, see
Atlas Vector Search.
Note
You cannot use a $vectorSearch
stage in a $facet
stage.
Learn More
To learn more about creating Atlas Vector Search indexes, see Index Vector Embeddings.
To learn more about
$vectorSearch
pipeline stage syntax and usage, see Vector Search Queries.