Hi SaiVihAI

Databricks Vector Search currently does not support creating multi-vector indexes that are directly based on multiple embedding columns in a Delta table. Creating separate vector indexes for each embedding column is an option.

Ref  How to create and query a vector search index | Databricks Documentation  Mosaic AI Vector Search | Databricks Documentation

Is this still the case? Both the rest api (https://docs.databricks.com/api/workspace/vectorsearchindexes/createindex) and the python sdk (https://databricks-sdk-py.readthedocs.io/en/latest/dbdataclasses/vectorsearch.html#databricks.sdk.se...) appear to suggest we should be able to pass multiple embedding_source_columns.