vectorIndexes

One or more vector indexes to be created on the table. Each vector index enables similarity search on a vector attribute. Each element in the list consists of:

• IndexName - The name of the vector index. Must be unique within the table.

• VectorAttribute - The attribute that contains vector embeddings. If multiple vector indexes reference the same attribute, they must all use the same number of dimensions.

• Dimensions - The number of dimensions in each vector.

• DistanceFunction - The distance function used to calculate similarity. Valid values: COSINE, EUCLIDEAN, DOT_PRODUCT.

• Projection - Specifies attributes that are copied (projected) from the table into the vector index. The total number of projected non-key attributes is shared across the vector attribute (counts as 1) and INLINE_FILTER search schema elements (each counts as 1). HASH search schema elements do not count toward this limit.

• SearchSchema - (Optional) Defines the partition key (HASH) and inline filter (INLINE_FILTER) attributes for the vector index.