Get Started

Lantern Cloud

The easiest way to get started with all of our tools is with Lantern Cloud.

In Lantern Cloud, you can create create a database with just a few clicks. Once you load your data into Lantern, you can generate embeddings with a single click from dozens of provided open source or proprietary embedding models. You can then create a vector index from your dashboard or run an index-tuning experiment to choose the best parameters for index creation.

We're currently in closed beta. Reach out at for access. We'd love to learn more about your use case and help out.


Alternatively, you can also use our tools locally or self-host them. There are three tools that are provided out-of-the-box with Lantern Cloud.

  • Lantern, our core Postgres extension, provides vector search in Postgres.
  • Lantern Extras, which further extends Postgres to support embedding generation.
  • Lantern CLI provides routines for generating embeddings and indexes.

You can install the tools individually by following the instructions linked.


Here is a non-comprehensive overview of what you can do with Lantern. The examples below use SQL, but we also provide examples for Python, JavaScript, and Rust.

Create a table with an embedding column


CREATE TABLE books (id SERIAL PRIMARY KEY, book_embedding REAL[3]);

Generate embeddings


SELECT text_embedding('BAAI/bge-base-en', 'My text input');
SELECT openai_embedding('openai/text-embedding-ada-002', 'My text input');

Insert embeddings


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

Calculate distance and select nearest rows


SELECT book_embedding <-> '{0,0,0}' FROM books
    ORDER BY book_embedding <-> '{0,0,0}' LIMIT 1;

Create an index to speed up nearest neighbor search


CREATE INDEX book_index ON books USING lantern_hnsw(book_embedding dist_l2sq_ops)
    WITH (M=2, ef_construction=10, ef=4, dim=3);