Introduction
Neon is a serverless Postgres platform designed for applications and agents. It provides a complete backend with database, authentication, storage, functions, and an AI gateway. Built on Lakebase Postgres, Neon offers instant, branchable, and scalable infrastructure. It is part of the Databricks Platform.
What is Neon?
Neon is a serverless Postgres database that separates storage and compute to enable autoscaling, branching, and instant operations. It includes additional backend services: Authentication (built on Better Auth), Object Storage (S3-compatible), Functions (serverless compute), and an AI Gateway (one API for frontier and open-source models). Neon is designed for developers and agents, with full access via MCP, CLI, and SDK.
What makes Neon stand out?
- Lakebase Architecture: Storage-compute separation for instant operations and scalability.
- Autoscaling: Monitors database load ten times per second and adjusts CPU and memory to fit workload.
- Branching: Create isolated copies of your entire backend for development, testing, and rollback.
- Instant Restore: Quickly recover from mistakes.
- Complete Backend Platform: Includes Authentication, Object Storage, Functions, and AI Gateway.
- Agent-Ready: Supports coding agents with instant operations and safe automation.
- Free Tier: 100 free projects with Database, Storage, Functions & Auth; 1 GB Postgres storage per project.
- Usage-based Pricing: Pay for what you use, with no fixed fees for security and compliance features.
- No-Fee AI Gateway: At-cost rates for frontier models with no add-on fees.
- Compliance: CCPA, GDPR, ISO 27001, ISO 27701, SOC 2, HIPAA compliant.
Who is it for?
- Startups: From bootstrapped to VC-funded, ship faster on Neon.
- Enterprises: Run critical workloads on Lakebase Postgres; trusted by 70% of Fortune 500.
- Agent Platforms: Deploy isolated backends for end users; used by Replit and v0.
- Developers: Build full-stack apps, simplify DB ops, and handle bursty workloads.
- AI Platforms: Deploy isolated backends for end users.
What can you do with Neon?
- Build full-stack apps: Deploy backends via your agent.
- Simplify DB ops: Use branching workflows to ship faster and safer.
- Handle bursty workloads: Avoid overprovisioning and optimize performance.
- Manage large databases: Restore and replicate your DB in seconds.
- Deploy at scale: Scale from first users to Fortune 500.
- Support agents: Provide infrastructure for app-generation agents.
- Deploy isolated backends: For AI platforms and end users.
How does Neon work?
Neon uses a Lakebase architecture that decouples storage and compute. This enables instant operations, autoscaling, and branching. Autoscaling monitors database load ten times per second and adjusts CPU and memory in real-time. Branching creates copy-on-write clones of your entire backend, allowing isolated environments for development, testing, and rollback. Neon provides full access via MCP, CLI, and SDK, making it suitable for automation by agents.
Pricing
Neon offers a Free & Paid pricing model. The Free Plan includes 1 GB of Postgres storage per project and 100 projects. Paid plans are usage-based: you pay for what you use, not for infrastructure you might need. There are no fixed fees for security and compliance features. The AI Gateway provides at-cost rates for frontier models with no add-on fees. For detailed pricing, visit the Neon pricing page.
