April 2026, Release Update 23.26.2

Included are some notable Oracle AI Vector Search updates with Oracle AI Database 26ai, Release Update 23.26.2.

Feature	Description
Function-Based Hierarchical Navigable Small World Indexes	HNSW vector indexes can now be created on an expression instead of a vector column. The index expression can be an arithmetic, conditional expression or any other expression that produces a vector. Function-Based Hierarchical Navigable Small World Index
Local Hierarchical Navigable Small World Indexes	HNSW vector indexes can now be created for each partition or sub partition of a partitioned table. Instead of building a single, global HNSW index across all vectors in a large, partitioned table, this enhancement enables the creation of individual HNSW indexes for each (sub)partition. Local Hierarchical Navigable Small World Index
Multi-Vector Search Using IVF Indexes	IVF indexes accelerate multi-vector search by partitioning the vector space into a configurable set of centroids. At query time, the search is limited to the most relevant partitions instead of scanning all vectors, delivering significantly faster response times for multi-vector workloads compared to a full scan. Multi-Vector Search Using IVF Indexes
IVF Vector Index on Iceberg External Tables	Oracle AI Vector Search now supports creating IVF vector indexes on Iceberg external tables stored in cloud object storage (OCI Object Storage, Amazon S3). IVF Indexing on External Iceberg Tables