pgvector
https://github.com/pgvector/pgvector
Stores embeddings alongside relational + full-text data. HNSW + IVFFlat ANN indexes. Used as agent conversation memory via LangChain + MCP. Foundation of Supabase AI and many self-hosted RAG stacks.
Active · Verified 2026-10-26 · fresh as of 2026-10-26 · 56 cited facts, 10 with a third-party source · 1 recorded change · Suggest a correction
At a glance
- Type
- PostgreSQL vector extension repo fix
- Tier
- T1
- Section
- Vector-database infrastructure
- Latest release
- v0.8.6 (2026-07-29) repo fix
- License
- PostgreSQL repo fix
- Pricing
- free (OSS) repo fix
- Freshness
- 2026-10-26 fix
- Layer
- model-infra repo fix
- SDK languages
- csharp, elixir, go, java, kotlin, php, python, ruby, rust, swift, typescript repo fix
- Developer activity
- OpenRank 2.89 (12-mo avg 3.53) · 16 participants — 2026-09 third-party fix
When to use
Optimised for: low-latency similarity search + scale
Anti-fit: not for relational / graph-heavy queries (vector-first by design)
Pros & cons
Pros
Postgres extension — keeps vector data alongside transactional data with full ACID; minimal ops for shops already running Postgres.
Cons
Performance plateau at very large vector counts; limited filter pushdown vs purpose-built engines.
Claims & capabilities
PostgreSQL license. Most-deployed open-source vector store.
Technical surface
- API surface
- SQL via Postgres protocol
- Backend storage
- Postgres
- Deployment
- Self-hosted only — PostgreSQL extension; also offered inside managed Postgres (Supabase, AWS RDS, Azure, Neon, etc.)
- Embedding model
- BYO
- Multi-tenancy
- Logical isolation via Postgres schemas/databases; managed providers (Supabase, Neon, RDS) add tenant-level controls
- MCP
- yes — community Postgres MCP servers with pgvector tools (e.g. MOmayr/mcp-postgres-pgvector)
- A2A
- not supported
- OpenTelemetry
- via Postgres OTel
Compare pgvector with…
Alternatives to pgvector → live options ranked by shared design, licence, hosting and MCP support
Similar systems
Other vector-database infrastructure in the catalog, ranked by inbound references.
- Qdrant T1
Distribution-Based Score Fusion + RRF. Sparse vectors native; filtering via ANN graph modification.
- Chroma T1
Limited native hybrid (users build RRF custom). Fast Rust core (v2.5). Acquired by Replit (definitive agreement announced 2026-10-23); Chroma Cloud is winding down (writes disabled 2026-11-22, reads disabled 2026-12-22) while open-source ChromaDB continues under Apache-2.0.
- Weaviate T1
Hybrid Search 2.0 with learned ranking. BM25 + vector in unified index.
- Milvus T1
Multi-vector columns (10 simultaneous). Native hybrid search (v2.5). CAGRA + Vamana GPU/CPU (v2.6).
- Pinecone T1
Managed vector DB. Cascading sparse + dense + rerank pipeline; pinecone-rerank-v0 .
- MongoDB Atlas Vector Search T1
Agent memory store for both short-term (document) and long-term (vector). LangGraph checkpointer for stateful agents. Vector search extended to Community Edition (Sept 2025).
Sources (55 cited facts, 10 third-party)
Each fact links to the page it came from; the badge says who published it (how to read sources).
- What it is PostgreSQL vector extension repo
- What it does & what's distinct Stores embeddings alongside relational + full-text data. HNSW + IVFFlat ANN indexes. Use… repo
- Claims / benefits PostgreSQL license. Most-deployed open-source vector store. repo
- Modalities text repo
- Created 2021-04 repo
- Latest release v0.8.6 (2026-07-29) repo
- License PostgreSQL repo
- GitHub pgvector/pgvector — ~23.5k stars, last commit 2026-07 or later (0.8.7 unreleased) repo
- Mindshare 3 inbound repo
- Funding No external funding — open-source project by independent developer repo
- Pricing free (OSS) repo
- Compliance Inherits PostgreSQL security model; no standalone vendor compliance repo
- Deployment Self-hosted only — PostgreSQL extension; also offered inside managed Postgres (Supabase,… repo
- Founders / pedigree Andrew Kane (sole maintainer/creator; first release April 2021) repo
- Memory volume / scale no scale claim repo
- Performance ANN-Benchmarks: pgvector HNSW achieves ~1k QPS at recall 0.95 on SIFT-1M; competitive wi… repo
- API surface SQL via Postgres protocol repo
- Backend storage Postgres repo
- Multi-tenancy Logical isolation via Postgres schemas/databases; managed providers (Supabase, Neon, RDS… repo
- Encryption At-rest + in-transit via underlying Postgres / managed-cloud encryption; CMEK via cloud … repo
- SSO / RBAC RBAC only (Postgres roles) repo
- Embedding model BYO repo
- Consistency strong third-party
- Versioning snapshots third-party
- Tombstoning hard-delete supported third-party
- Schema evolution schema with migration third-party
- Namespace primitives per-tenant third-party
- Contradiction handling always overwrite (last-write-wins) third-party
- Forgetting policy user-controlled forget third-party
- MCP support yes — community Postgres MCP servers with pgvector tools (e.g. MOmayr/mcp-postgres-pgvec… repo
- A2A support not supported repo
- OTel via Postgres OTel third-party
- Webhooks / events Postgres LISTEN/NOTIFY third-party
- Import / export SQL dump, CSV, COPY repo
- Runtimes Client libs for 35 languages incl. Python, JavaScript/TypeScript, Go, Java, Rust, C#, Ru… repo
- Validated verticals RAG developers, agent builders (broad horizontal); financial-services, healthcare, legal… repo
- Time to running <10min (managed cloud) to <1hr (self-host Docker) repo
- Anti-fit not for relational / graph-heavy queries (vector-first by design) repo
- Optimised for low-latency similarity search + scale repo
- Adjacent infrastructure BYO embedding model + LLM; this IS the vector store repo
- Pros Postgres extension — keeps vector data alongside transactional data with full ACID; mini… repo
- Cons Performance plateau at very large vector counts; limited filter pushdown vs purpose-buil… repo
- Project & sources GitHub pgvector repo
- commit-trajectory [{"ym":"2023-09","cum":798},{"ym":"2023-10","cum":826},{"ym":"2023-11","cum":862},{"ym":… repo
- download-trajectory [{"ym":"2025-11","monthly":5644492,"cum":5644492},{"ym":"2025-12","monthly":11929244,"cu… registry
- cost-input-usd-per-mtok 0 repo
- cost-output-usd-per-mtok 0 repo
- cost-tier free repo
- cost-pricing-model free-tier repo
- cost-last-verified 2026-06-29 repo
- architecture-deployment-locus local repo
- positioning-platform-vs-product platform repo
- taxonomy-three-layer-rollup model-infra repo
- developer-activity-opendigger {"activity":29.8,"as_of":"2026-09","openrank_12mo":3.53,"openrank_current":2.89,"partici… third-party
- sdk-languages csharp,elixir,go,java,kotlin,php,python,ruby,rust,swift,typescript repo
What changed
Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)
- 2026-10-26
- License “Custom” → “PostgreSQL” source
- Latest release first recorded as “v0.8.6 (2026-07-29)” source
- Pricing “Free (PostgreSQL License open source)” → “free (OSS)” source
- Deployment “Self-hosted only (PostgreSQL extension; managed via Supabase/AWS RDS/etc.)” → “Self-hosted only — PostgreSQL extension; also offered inside managed Postgres (Supabase, AWS RDS, Azure, Neon, etc.)” source
- MCP support “via Postgres MCP servers (community)” → “yes — community Postgres MCP servers with pgvector tools (e.g. MOmayr/mcp-postgres-pgvector)” source
Related systems
References (1)
- LangChain (framework) integrates with · depends on at runtime — Used as agent conversation memory via LangChain + MCP.
Referenced by (17)
- Agno (Phidata) Memory builds on — Postgres + pgvector default — operational simplicity for shops that already run Postgres.
- Apache AGE integrates with — openCypher over Postgres. Pairs with pgvector for graph + vector hybrid retrieval.
- Dify integrates with — Dify self-host VECTOR_STORE supports `pgvector` as the knowledge-base vector store. (verified 2026-10-29)
- Haystack (framework) integrates with — Haystack integrations directory has a pgvector integration page ("A Document Store for storing and retrieval from pgvector"). (verified 2026-10-29)
- LangChain (framework) integrates with — LangChain docs have a PGVector vector store integration page (langchain-postgres). (verified 2026-10-29)
- Langflow integrates with — Langflow components bundle: PGVector vector store component. (verified 2026-10-29)
- Letta / MemGPT depends on at runtime — backend-storage cell: Postgres + pgvector
- LlamaIndex (framework) integrates with — LlamaIndex has an official vector store integration package llama-index-vector-stores-postgres in its repo. (verified 2026-10-29)
- Mastra integrates with — Mastra repo ships the @mastra/pg package (PgVector, vector search using pgvector). (verified 2026-10-29)
- Mem0 integrates with — Mem0 docs have a pgvector vector store page for Mem0 open source. (verified 2026-10-29)
- Mem0 Security / OpenMemory builds on · depends on at runtime — For privacy-first deployments the self-hosted path is now the Mem0 open-source server (REST API + dashboard via Docker Compose with pgvector Postgres, admin setup wizard, per-user API keys, request…
- Memobase depends on at runtime — backend-storage cell: Postgres + pgvector
- Microsoft Semantic Kernel integrates with — Semantic Kernel Postgres vector store connector docs say it uses the pgvector extension. (verified 2026-10-29)
- n8n (AI agent flows) integrates with — n8n built-in AI cluster node: PGVector Vector Store node. (verified 2026-10-29)
- OpenLLMetry (Traceloop) integrates with — OpenLLMetry's What's Supported page lists pgvector as instrumented out of the box. (verified 2026-10-29)
- Phidata (now Agno) integrates with — Agno ships a pgvector vector DB integration module (agno.vectordb.pgvector). (verified 2026-10-29)
- Supabase Vector integrates with — Supabase AI & Vectors: vector store and embeddings support using Postgres and pgvector. (verified 2026-10-29)