Maintained with ☕️ by
IcePanel logo

Spanner vector index and approximate nearest neighbor (ANN) distance functions in the GoogleSQL-dialect are Generally Available

Share

Services

## Feature Spanner vector index and approximate nearest neighbor (ANN) distance functions in the GoogleSQL-dialect are Generally Available. If you have a table with a large amount of vector data, you can use a vector index to accelerate similarity searches and nearest neighbor queries. Spanner now also supports the following: * [ALTER VECTOR INDEX](https://cloud.google.com/spanner/docs/reference/standard-sql/data-definition-language#alter-vector-index) DDL syntax * [Import and export](https://cloud.google.com/spanner/docs/import-export-overview) databases that use ANN * Use the [STORING clause](https://cloud.google.com/spanner/docs/secondary-indexes#storing%5Fclause) to store a copy of a column in the vector index to accelerate queries that filter by those columns * Use ANN in instances smaller than one node or 1000 processing units For more information, see [Find approximate nearest neighbors, create vector indexes, and query vector embeddings](https://cloud.google.com/spanner/docs/find-approximate-nearest-neighbors). ## Feature Spanner ANN indexes are now supported in Langchain. For more information, see [LangChain Quickstart for Spanner](https://github.com/googleapis/langchain-google-spanner-python/blob/main/samples/langchain%5Fquick%5Fstart.ipynb).