pgvectorscale adds a StreamingDiskANN index to pgvector, letting PostgreSQL handle very large vector collections at high speed.
| Category | Vector database |
| Type | PostgreSQL extension |
| License | PostgreSQL |
| Runs locally | Yes |
| Built with | Rust |
| Skill level | Intermediate |
| Best for | scaling pgvector past a few million rows |
Other open-source vector database tools worth comparing:
QdrantFast Rust-based vector search
WeaviateVector DB with built-in modules
MilvusBillion-scale vector database
ChromaThe simplest embedding DB for prototyping
pgvectorVector search inside PostgreSQL
LanceDBServerless vector DB for AI
FAISSThe reference library for similarity search
VespaBig-scale hybrid search and ranking platform
MarqoVector search with embedding built in
USearchTiny, extremely fast vector indexpgvectorscale is free and open-source (PostgreSQL license), so you can use, self-host and modify it at no cost.
Yes. pgvectorscale is designed to run on your own machine or server, keeping your data private.
Popular open-source alternatives include Qdrant, Weaviate, Milvus. See the comparisons above to choose.
Browse the full directory of open-source AI tools, models and projects — updated daily.
Browse all tools →