Lantern Howto

Sign up for Lantern Cloud: Go to lantern.dev and click 'Try Lantern for Free' to create a free account. No credit card is required.

Create a database: After signing up, create a new Postgres database with Lantern enabled.

Connect to your database: Use the provided connection details to connect to your Lantern-enabled Postgres database using your preferred method (e.g. psql, application code, etc.).

Create a table with a vector column: Execute SQL to create a table that includes a column for storing vector embeddings, e.g. 'CREATE TABLE books (id SERIAL PRIMARY KEY, book_embedding REAL[3]);'

Insert vector data: Insert vector embeddings into your table, e.g. 'INSERT INTO books (book_embedding) VALUES ('{0,1,0}'), ('{3,2,4}');'

Create an HNSW index: Create a Lantern HNSW index on your vector column for faster queries, e.g. 'CREATE INDEX book_index ON books USING lantern_hnsw(book_embedding dist_l2sq_ops) WITH (M=2, ef_construction=10, ef=4, dim=3);'

Perform vector similarity search: Use SQL to query for similar vectors, e.g. 'SELECT id FROM books ORDER BY book_embedding <-> '{0,0,0}' LIMIT 1;'

Generate embeddings (optional): Use Lantern's built-in embedding generation to create vectors from text or images, e.g. 'SELECT id FROM books ORDER BY book_embedding <-> text_embedding('BAAI/bge-base-en', 'My text input') LIMIT 1;'