Skip to content

feat: Add documentation for CREATE VECTOR INDEX (#27332) #27332

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
skyelves merged 1 commit into from
Mar 20, 2026
+53 −0
Merged

feat: Add documentation for CREATE VECTOR INDEX (#27332) #27332

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions presto-docs/src/main/sphinx/sql.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@ This chapter describes the SQL syntax used in Presto.
sql/create-schema
sql/create-table
sql/create-table-as
sql/create-vector-index
sql/create-view
sql/create-materialized-view
sql/deallocate-prepare
Expand Down
52 changes: 52 additions & 0 deletions presto-docs/src/main/sphinx/sql/create-vector-index.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,52 @@
===================
CREATE VECTOR INDEX
===================

Synopsis
--------

.. code-block:: none

CREATE VECTOR INDEX index_name
ON source_table ( column_name [, ...] )
[ WITH ( property_name = expression [, ...] ) ]
[ UPDATING FOR predicate ]

Description
-----------

Create a vector search index on the specified columns of a source table.

The ``column_name`` list specifies the columns to index. Typically this includes
an identifier column and an embedding (vector) column.

The optional ``WITH`` clause sets index properties such as:

- ``index_type``: The type of vector index (for example, ``'ivf_rabitq4'``)
- ``distance_metric``: The distance metric to use (for example, ``'cosine'``)
- ``index_options``: Additional index configuration (for example, ``'nlist=100000,nb=8'``)
- ``partitioned_by``: Partition columns (for example, ``ARRAY['ds']``)

The optional ``UPDATING FOR`` clause filters the source data using a predicate,
similar to a ``WHERE`` clause. This is typically used to specify which partitions
to build the index for.

Examples
--------

Create a vector index on an embeddings table for specific date partitions::

CREATE VECTOR INDEX my_index
ON my_table(content_id, embedding)
WITH (
index_type = 'ivf_rabitq4',
distance_metric = 'cosine',
index_options = 'nlist=100000,nb=8',
partitioned_by = ARRAY['ds']
)
UPDATING FOR ds BETWEEN '2024-01-01' AND '2024-01-03'

See Also
--------

:doc:`create-table`, :doc:`create-table-as`
Loading