Scalable, fast, and disk-friendly vector search in Postgres, the successor of pgvecto.rs.

github.com/tensorchord/VectorChord
VectorChord homepage screenshot

About VectorChord

VectorChord is a PostgreSQL extension for vector search, distributed as an open-source-style repository on GitHub. Its homepage describes it as a successor to pgvecto.rs and aims at scalable, high-performance, and cost-effective vector search. It stores vectors using RaBitQ compression with reranking, and offers native 4-bit and 8-bit vector types for lower storage use.

The extension is designed to work alongside pgvector. The homepage states that it is fully compatible with pgvector data types and syntax, and that it depends on pgvector for the vector representation. A Docker image is provided for quick local setup, along with an all-in-one image that includes VectorChord-bm25 and pg_tokenizer.rs. Performance and cost comparisons on the homepage are presented by the project itself.

Key features

  • Vector search implemented as a PostgreSQL extension, compatible with pgvector data types and syntax
  • RaBitQ compression with autonomous reranking to reduce vector storage size
  • Native 4-bit and 8-bit vector types for lower storage use
  • vchordrq index type for vector similarity search queries
  • Hierarchical K-means index building, which the homepage says can index 100 million vectors in 20 minutes
  • Docker images for quick setup, including a base image and an all-in-one image
Best for: Teams already running PostgreSQL who want vector search inside their existing database and are concerned with storage size and infrastructure cost at large scale.
Good for
  • Compatible with pgvector data types and syntax, so existing pgvector applications can be migrated with little change
  • Provides low-bit vector types, with the homepage stating that RaBitQ8 has under 1% recall loss
  • Includes a Docker image that allows a quick start without a manual installation
Watch out for
  • Depends on pgvector, which must be present for the vector representation
  • Performance and cost figures on the homepage are the project's own comparisons and are not independently verified in the source

VectorChord at a glance

Type
Open-source project
Pricing
Not confirmed. Check the site
Open source
Not stated
Language
English
Last checked
Oct 6, 2026 · Online when checked
Listed
Oct 5, 2026

VectorChord alternatives

Tools and sites you could use instead, ranked by how closely they match.

All 24 alternatives

Popular in AI Infrastructure & MLOps

Well-known listings from the same topic.

All 341

VectorChord FAQ

Is VectorChord compatible with pgvector?
The homepage states that VectorChord is fully compatible with pgvector data types and syntax. It also depends on pgvector for the vector representation.
How can I start VectorChord quickly?
The homepage recommends the Docker image ghcr.io/tensorchord/vchord-postgres. After connecting to the database with psql, you can run CREATE EXTENSION IF NOT EXISTS vchord CASCADE;.
Which image includes all official TensorChord extensions?
The homepage says the image tensorchord/vchord-suite:pg17-latest includes VectorChord, VectorChord-bm25 and pg_tokenizer.rs. Developers should choose a tag compatible with their extension version, as shown in the support matrix.
Is something wrong or out of date?

Tell us and an editor will check it.

Own this website?

Claim the listing and suggest updates yourself. An editor checks every claim first.

Claim this listing