tensorchord/pgvecto.rs

Health v2 · maintenance only

Adoption not part of the score

2183 stars · 86 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

pgvecto.rs is a Postgres extension written in Rust (via pgrx) that adds vector similarity search with high dimension support, quantization, and filtered vector queries. It is now superseded by its successor VectorChord, which the maintainers recommend users migrate to.

Use cases

• add vector similarity search to an existing postgres database
• store and query llm embeddings without running a separate vector database
• run filtered nearest-neighbor search with joins in sql
• build a rag application backed by postgres
• index high-dimensional embeddings up to 65535 dimensions
• compare pgvector vs pgvecto.rs performance for vector search

When to choose

• you already run postgres and want vector search without a second database
• you need vector dimensions beyond pgvector's 2000 limit
• you need combined vector search with relational filters and joins
• you want quantization options to reduce memory usage

When to avoid

• starting a new project - the maintainers recommend VectorChord instead
• you need full WAL support for indexes
• you want a managed or standalone specialized vector database
• you need a battle-tested, stable extension for critical production workloads

Facets

plugin · maturity maintenance

vector-database search-engine database databases machine-learning large-language-models self-hosted rust python postgres-extension vector-similarity-search hnsw embeddings pgvector-alternative pgrx simd hybrid-search search retrieval-augmented-generation linux docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="tensorchord/pgvecto.rs")