Introducing pgvector - Perform Semantic Search on a Postgres database

Introducing pgvector - Perform Semantic Search on a Postgres database

The kind folks at Supabase recently released pgvector - an extension that enables you to perform vector search fo’r your AI/ML applications on top of a Postgres database.

Can I install pgvector on Managed Databases?

Unfortunately, popular providers of Managed Postgres such as AWS RDS, Google and Azure do not enable you to install custom extensions.

Where Can I install pgvector on a Managed Database?

Fortunately, there are serverles Postgres Providers that support installation of custom extensions
  1. Supabase
notion image
Of course, the makers of this package also allow installation on their managed service.
notion image supports most but not every PostgreSQL construct, and numerous extensions. You can find the list of supported extensions here.
  1. Crunchy Bridge
notion image
Crunchy Bridge is a fully managed Postgres service from Crunchy Data that works across every major cloud provider — Amazon Web Services, Google Cloud Platform, and Microsoft Azure. The full catalog of Crunchy Bridge’s extensions and languages is available here.


Embeddings are succinct representations of complex data in the form of vectors enabling you to capture semantic similarity among elements. With the recent rise of Large Language Models from companies like OpenAI and Cohere, sufficiently complex semantics can be captured by these vectors. Go deeper on vector embedding on this guide here.

About NNext

At NNext, we are building the tools enabling you to run semantic search in familiar SQL semantics. If you’d like early access, register on our app at!