supervc-stack/VectorChord

Health v2 · maintenance only

Flags: no_license

Adoption not part of the score

1807 stars · 72 forks observed · 2026-09-30

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

VectorChord is a PostgreSQL extension written in Rust for scalable, disk-efficient vector similarity search, serving as the successor to pgvecto.rs. It uses RaBitQ quantization with autonomous reranking to store hundreds of millions of vectors on modest hardware while remaining pgvector-compatible.

Use cases

• run vector similarity search inside postgres
• store 100 million embeddings on a single server
• build a RAG pipeline without a separate vector database
• migrate from pgvector for better performance and lower storage cost
• do hybrid keyword and vector search in one database
• index large embedding datasets quickly
• support high-dimensional embeddings like text-embedding-3-large

When to choose

• you already run PostgreSQL and want vector search without adding a dedicated vector database
• you need billion-scale vector search at low infrastructure cost
• you want pgvector-compatible syntax with faster queries and indexing
• you need long vectors up to 60,000 dimensions
• you want BM25 plus vector search together for RAG

When to avoid

• you need a fully managed vector database service
• your project is not on PostgreSQL
• you only have a few thousand vectors and pgvector is sufficient
• you require a permissive open-source license without restriction

Facets

library · maturity active

vector-database search-engine database databases machine-learning large-language-models windows self-hosted cloud postgresql-extension vector-search pgvector-compatible rabitq-quantization rust rag similarity-search embeddings search retrieval-augmented-generation linux macos docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="supervc-stack/VectorChord")