Neon now has realtime: Live SQL with full reactivity. Try it in Beta today
Lakebase Postgres

The Neon database: Lakebase Postgres

Built on the lakebase architecture. Decoupled storage and compute, with object storage + WAL as the foundation.

Deploy, scale, branch, replicate, and restore instantly, without moving or duplicating your underlying data between environments.

  • Ephemeral compute

    Provisions instantly, autoscales with load, and scales to zero when idle, restarting in <1s.

  • Shared storage

    The same versioned storage built on WAL serves every branch and replica in a project.

  • Agent-ready

    Operations are lightweight and run through the API, CLI, SDKs, and MCP.

Autoscaling

Compute follows your traffic. No overprovisioning or performance hiccups.

There’s no instance to size and no manual resizes in Lakebase Postgres: your database autoscales in real time

outages prevented by Autoscaling this year

  • Neon autoscaling

  • Database load

  • Fixed-resource provisioned

Neon monitors your database load ten times a second and autoscales CPU and memory to exactly fit your workload.

  1. Scales under load

    Compute moves within your set range, up during spikes and down during slow times.

  2. Suspends when idle

    After five minutes without activity (e.g. in dev environments), compute scales to zero.

  3. Only bills for what runs

    You pay for the compute you actually use, without having to worry about sizing.

Agents demand new database primitives like branching, together with instant deploys and restores, and full CLI/MCP coverage.

Create a full copy of production in about a second, without duplicating storage, so a 1 TB branch takes just as long to create as a 1 GB branch.

Production branches into isolated checkout-test and payment-fix environments, each with its own database, functions, and authentication.
  • An environment per unit of work

    Every PR, version, test, and preview can have its own backend branch, following your code.

  • Fully programmable

    Creating and deleting branches is a lightweight metadata operation: your agent and the API can manage it end to end.

  • The whole backend branches

    A Neon branch also carries its own Object Storage namespace, its own Functions, its own AI Gateway endpoint, and its own auth.

Roll back instantly to any point in your database history, without copying data, so restore time stays small no matter how large your database is.

A database timeline rolls back from a query error to a checkpoint, creating a restored main branch.

Let agents deploy and operate isolated database environments for every task, session, or pull request, with full isolation and undos always at hand.

Data API

Query Postgres directly from browsers, edge runtimes, and serverless functions.

data-api.ts
await fetch(`${DATA_API_URL}/projects`, {
  method: "POST",
  headers: {
    Authorization: `Bearer ${token}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({
    name: "New project"
  })
})
  • Instant REST API

    Turn Postgres tables, views, and functions into REST endpoints you can access directly over HTTPS.

  • Secure access

    Authenticate with JWTs and use Postgres Row-Level Security to control who can access and modify your data.

  • PostgREST compatible

    Bring any PostgREST client, including @neondatabase/postgrest-js, and keep familiar filtering, ordering, pagination, and CRUD patterns.

From your first five users
to the world's largest teams

From early-stage startups to Fortune 500 organizations, Lakebase Postgres gives you the same flexible foundation to build, scale, and run production workloads with confidence.

    Ship faster with a small team

    Adopt branching workflows and grow your startup faster. Don’t let the database lag the speed at which you ship code.

    Retool
    “We’ve been able to manage 300K+ Postgres databases via the Neon API. It saved us a tremendous amount of time and engineering effort.”
    Himanshu Bhandoh – Software Engineer at Retool
    Read case study

    Your questions,
    answered.

    • In terms of compatibility, yes. Lakebase Postgres is Postgres — your existing drivers, ORMs, migration tools, and everything else from the Postgres ecosystem works unchanged. What’s different is the architecture underneath.

    Your Postgres branches with everything else. Create a branch and your whole backend forks with it — database, storage, auth, and a gateway endpoint of its own.

    • Lakebase Postgres. Serverless Postgres that scales and branches with your app.

    • Authentication. Managed auth with users and sessions stored in Postgres.

    • Compute. Functions without timeouts running close to your database.

    • Storage. S3-compatible object storage that branches with your projects.

    • AI Gateway. One API for all frontier & open-source models, powered by Databricks.

    Backed by giants

    Trusted at scale. Neon has been part of the Databricks Platform since May 2025.

    • 20M+

      Databases started daily - built for scale and reliability.

    • 3M+

      Developers building on Neon worldwide

    Trusted by the best
    Neon's serverless philosophy is aligned with our vision: no infrastructure to manage, no servers to provision, no database cluster to maintain.
    Edouard Bonlieu Co-founder at Koyeb
    Get help

    Talk to us.

    Fill out a short form and we’ll get back to you within a few business days.

    Contact us