A vector database that's also a database.
pgvector in every project. Store embeddings next to relational data, search semantically and build RAG without syncing to a separate vector store.
create extension if not exists vector;
create table documents (
id bigserial primary key,
content text,
embedding vector(1536)
);
create index on documents using hnsw (embedding vector_cosine_ops);
select content from documents
order by embedding <=> $1
limit 5;Capabilities
What you get with Vector
pgvector and HNSW
HNSW and IVFFlat indexes for fast search across millions of vectors.
Hybrid search
Combine vector similarity with full-text search and SQL filters.
Any model
OpenAI, Anthropic, Cohere, Mistral or open-source models — it's just a vector column.
RLS for embeddings
Users only search their own documents — no extra code.
Automatic embeddings
Triggers and Edge Functions that generate vectors when a row is written.
MCP server
AI agents query your database through the Model Context Protocol.
Use cases
What teams use it for
Chatbots with company knowledge
RAG over docs, policies and your knowledge base.
Semantic search
Results by meaning, not just keywords.
Recommendations
Similar products and content based on embeddings.
The rest of the platform
Every service shares the same database, authentication and permissions.
Build in a weekend. Scale to millions.
Spin up a free project in 2 minutes. No credit card, no pausing after a week.