Postgres continues to evolve the database landscape beyond traditional relational database use cases. Its rich ecosystem of extensions and derived solutions has made Postgres a formidable force, especially in areas such as time-series and geospatial, and most recently, gen(erative) AI workloads.

Pgvector has become a foundational extension for gen AI apps that want to use Postgres as a vector database. In brief, pgvector adds a new data type, operators, and index types to work with vectorized data (embeddings) in Postgres. This allows you to use the database for similarity searches over embeddings.