YugabyteDB 2025.2 – Ultra-resilient scalable distributed PostgreSQL AI database

Summary: YugabyteDB 2025.2 supports vector data storage and similarity search using HNSW indexing, scaling to billions of vectors with high recall and low latency. It enables storing and querying OLTP and vector data together via SQL joins.

What it does

It provides a vector data store with HNSW-based indexing for similarity search, handling large-scale vector datasets alongside enterprise OLTP data in one database.

Who it's for

Users needing integrated storage and querying of vector and transactional data for AI and RAG applications.

Why it matters

It reduces total ownership cost by enabling efficient, combined vector and OLTP data management with simple SQL queries.