Vector Engine

Suggest edits

Vector Engine is a standalone PostgreSQL extension suite — it installs directly into your database and runs without AI Factory or Hybrid Manager. There is no platform dependency, no agent UI required, and no managed infrastructure needed. Everything operates through standard SQL.

It enables you to store, index, and search complex data like text or images by transforming them into mathematical coordinates, powering Retrieval-Augmented Generation (RAG) and semantic search applications directly inside your Postgres database.

Note

AI Factory is a separate EDB platform that sits above Vector Engine. It adds model serving, an agent UI, and managed orchestration for teams that want those capabilities. Vector Engine works independently of it — if you're looking for SQL-based embedding and search on your own Postgres instance, you're in the right place.

AIDB

AIDB acts as the orchestrator of your AI workflows. It automates the complex bakend tasks required to make your data AI-ready:

  • In-Database LLM integration: Connect directly to OpenAI, Azure, Google Cloud Storage, AWS S3, or local models using SQL.

  • AI data preparation: Automatically transform table data into vector embeddings.

  • Semantic management: No more glue code, you can manage your RAG workflows with standard SQL commands.

For more information, see AIDB docs

Vector Indexes

We offer a range of specialized indexing technologies to match your specific performance and accuracy needs:

  • pgvector: The industry standard for robust, reliable vector similarity search (IVFFlat and HNSW).

  • vectorchord: A cutting-edge index optimized for ultra-fast, low latency and high-throughput vector retrieval.

  • vectorchord-bm25: Enables hybrid search, combining the best of semantic vector retrieval with traditional keyword searching.

← Prev

Benchmark datasets

↑ Up

EDB Postgres AI Home

Next →

EDB Postgres AI Migration and ETL

Could this page be better? Report a problem or suggest an addition!