Build a vector database with one click

Lantern Cloud is a hosted Postgres vector database and toolkit for developers to build high-performance AI applications.

Try Lantern for Free
Backed by Combinator

Easy to use — just use SQL or your favorite ORM

Perform vector generation and vector search using SQL or a client library of your choice.
-- Create a table with a vector column
CREATE TABLE books (id SERIAL PRIMARY KEY, book_embedding REAL[3]);

-- Insert some vectors
INSERT INTO books (book_embedding) VALUES ('{0,1,0}'), ('{3,2,4}');

-- Create an index for faster queries
CREATE INDEX book_index ON books USING lantern_hnsw(book_embedding dist_l2sq_ops)
  WITH (M=2, ef_construction=10, ef=4, dim=3);

-- Query the nearest vector
SELECT id FROM books ORDER BY book_embedding <-> '{0,0,0}' LIMIT 1;

-- Query the nearest vector to a text embedding
SELECT id FROM books ORDER BY book_embedding <-> text_embedding('BAAI/bge-base-en', 'My text input') LIMIT 1;

One click vector generation

Lantern makes it easy to add vectors to your application.
Vector generation demo
20+ embedding models supported
Including Open AI, Cohere, Jina AI, and other open source embedding models.
Generate up to 2 million embeddings per hour.
Easy to use
1 click vector generation. Always up to date vectors.

Leverage unstructured data with the most powerful vector database

Lantern index creation is 30x faster than pgvector and can be 94% cheaper than Pinecone*.

HNSW Params m: 16, ef_construction: 128, sift 1M dataset, 8vCPU

Lantern offers best-in-class throughput, latency, and index creation times to handle the most demanding applications at scale.

Learn how

Save 20x-30x on cloud costs over competing vector databases

Customers switch to Lantern to avoid outrageous cloud costs generated by vector database competitors. Tap into industry leading performance for a fraction of the price

Pinecone logo

1 billion embeddings



Milvus logo

1 billion embeddings



1 billion embeddings



Assumes 1536 dimensional vectors and analogous price performance options like Pinecone's P2 performance-optimized and Milvus Enterprise metrics based on publicly available data for 100M embeddings (~$36,000/mo)