platoseedplatoseed
Postgres vector database extension to build AI applications
Lantern is the easiest way to build AI applications using Postgres. We started Lantern because we believe AI is going to impact every single enterprise in every single industry. We want to allow every company to tap into their unstructured data to build better applications. With Lantern Cloud, developers have access to everything they need to build an AI application: embedding generation, vector compression, vector search, efficient indexing, and more. All on top of the database they already know, Postgres, and using infrastructure that scales to billions. Check out our code on Github: https://github.com/lanterndata/lantern
Lantern is an open-source Postgres vector database and toolkit designed to help developers build AI applications directly inside Postgres. It offers vector search, BM25 text search, serverless indexing, and embedding/LLM integration within the database, with a managed cloud option available.
Lantern provides a Postgres extension and toolkit for vector search and AI workloads. It supports dense and sparse vector search, binary, scalar, and product compression, serverless indexing to scale indexes separately from the main database, and text search with BM25. It enables embedding generation and running LLMs inside Postgres, with support for multiple embedding models (OpenAI, Cohere, Jina AI, and open-source options). It offers a self-hosted open-source option and a managed Lantern Cloud service, with easy SQL-based usage and ORM integrations. The pricing page shows a free tier and paid tiers with serverless indexing and per-instance pricing, plus per-GB embedding and per-model costs for generation.
Pricing page with production tiers; open-source and cloud offering; documented embeddings/LLM integrations; user testimonials on scaling and search performance.
I'm working on Lantern. We're building the best platform to develop AI applications. Before Lantern, I worked at Y Combinator, on my own startups, and at Facebook. I studied computer science at Princeton, and grew up in Americus, Georgia.
Now building the only database you'll need for your AI applications. I was an early engineer at Timescale, a billion-dollar Postgres company, where I contributed core features. I did my undergrad at Princeton, where I built a safe OS schedule in Rust for my thesis. More recently, I was a PhD Student at UC Berkeley working on distributed systems.
Lantern is a Postgres vector database that is scalable, cost-effective, and easy to use
Lantern is a Postgres vector database that is easy to use, scales to billions, and is cheaper than standalone vector services. It supports embedding generation inside Postgres, product quantization for memory and cost savings, external index creation, and provides a managed service via Lantern Cloud with migration tools for existing vector or Postgres users.
From the original launch (Mar 2024) — may be outdated.
Open-source, serverless vectordb for production-scale generative AI
PandasAI is an open-source conversational data analysis platform with…
