Comparing token economics across 42 AI models
If headcount is often a company's largest cost, tokens will be a close second soon as agents take on more work. Imagine how much $$ a software company will be able to save just by pre-selecting model…
Blog — Neon Docs
The latest product updates from Neon
September 2026
Give your agent Neon tools
Today we're launching @neon/tools, a package that turns the @neon/sdk ergonomic client into typed agent tools with adapters for MCP, Mastra, and Eve. We're also using it to expand the hosted Neon MCP…
Building patterns unlocked by scale to zero
Most hosted Postgres pricing works like this: you pick an instance size, the vendor assumes the instance will run for about 730 hours a month, and you pay for those hours. Lakebase Postgres addresses…
Neon is now available in Grok Bot
Neon is now available as a plugin in Grok Bot. You just have to connect your Neon account once and you'll be able to ask a Grok bot to build on Neon.
August 2026
Autoscaling Lakebase Postgres
Choosing a database instance size before you know the workload is an old building pattern. Lakebase Postgres omits the sizing experience altogether thanks to autoscaling.
Introducing Neon Labs
Today we're launching Neon Labs, a home for experimental tools built around Lakebase Postgres. We want Neon Labs to be a playground for ideas that could help the broader Postgres community.
Just landed in the Neon CLI
The Neon CLI moved fast over the last couple of months. It is a key tool for agents, who do all the work without ever opening the Console - let’s take a look at some of recent changes.
WAL + S3: Lakebase storage for the era of agents
Agents treat code as lightweight: branch it, deploy it, throw it away. The infrastructure under that code should work the same way. Working with an OLTP database is traditionally heavy and full of fr…
Open-weight models are fast on Neon AI Gateway. Here's why
Open-weight models are built for performance, but how they're served determines whether end users actually feel that speed. Neon AI Gateway calls Databricks Foundation Model APIs for Databricks-hoste…
Query Neon backend logs
We keep expanding our backend observability - the most recent addition: you can now query backend logs outside the Console! neon logs reads what Neon Functions and Object Storage emit on a branch, wi…
LLMs belong in your backend
Neon AI Gateway is how we put LLM calls into the Neon backend the same way we put files, functions, and auth there. It allows you to call models right from Neon, instead of collecting lab accounts.
Neon skills landed in the Vercel CLI
Vercel just gave their CLI a new trick: when you install a Marketplace integration from the terminal, it also installs that provider's agent skills. A single command now provisions a Neon database on…
Neon for Herdr: A Neon pane in your Herdr workspace
I just released Neon for Herdr, a Herdr plugin that puts your Neon projects in a pane next to whatever you are already doing, so creating a branch, grabbing its connection string, or suspending a com…
Neon Functions: backend logic next to your data
Neon Functions are Node.js 24 compute you deploy onto a Neon branch, in the same region as your Lakebase Postgres database, with DATABASE_URL injected automatically. They're long-running enough that…
Electric is joining team Neon at Databricks
Today Databricks announced that Electric, the team behind PGlite and the Electric sync engine, is joining Databricks. We have our own reason to celebrate: Electric is joining the Neon team within Dat…
Building Neon Object Storage
We just launched Neon Object Storage, a branch-aware, S3-compatible object store built into the Neon backend. If you've used Neon branches before, you know the workflow: branch main into a child and…
Training 100x Cheaper Retrieval models Neon and Castform
“Most teams' best training data is just sitting in their databases. The problem is that turning raw data into something usable is hard, and letting agents read, search, and mutate data cheaply at sca…
Neon now has per-project permissions
Neon now supports a new permission structure, with four org-level roles (Admin, Editor, Viewer, and Collaborator) and per-project permissions that let you control exactly which projects your agents a…
July 2026
CommSync Runs Text, Vector, and Hybrid Search on Postgres with Lakebase Search
"We're running lakebase_text and lakebase_vector across three different parts of our product. Query latency improved 1,000x and we don't need a separate search stack sitting next to our Postgres."
My agent cloned GitHub (sort of) using the Neon backend
GitHub has been down a lot lately. During one of those episodes, I had the idea of asking my agent to use our new backend suite (still in beta) to build a working Git host. This is genuinely outside…
Debug your Postgres from the terminal: a tour of `neon inspect db`
The Neon CLI now ships `neon inspect db`: read-only, high-signal Postgres diagnostics that answer "what is slow?" without leaving the terminal or writing a single catalog query. Each subcommand runs…
Test out the Neon backend: Object Storage, Functions, and AI Gateway are beta
Storage, Functions, and AI Gateway are now in beta and available to all users: object storage that branches with your data, long-running compute that lives next to your database, and one API with one…
Introducing @neon/sdk, our new TypeScript client for the Neon API
The Neon API is open and follows the OpenAPI spec, so you can provision and manage almost everything you see in the console programmatically. But provisioning infrastructure over REST means polling o…
Introducing the New Neon API Command
CLIs have become the default tool for AI agents because they package complex environments into predictable commands. But human-friendly CLIs lag behind the underlying API, and letting agents call the…
Provision Your Entire Stack with Neon and Stripe Projects Build
A few months ago, Stripe changed how we think about backend infrastructure by launching Stripe Projects. Today, Stripe is taking it one step further with `stripe projects build`. Discover and scaffol…
Lakebase Search: vector and BM25 on Neon
Today we're shipping the beta of Lakebase Search on Neon: hybrid vector + full-text retrieval via two Postgres extensions, lakebase_vector and lakebase_text. This post is the story of why and how we…
June 2026
Introducing neon.ts: infrastructure as code for your Neon projects
As Neon turns into a platform with backend primitives for apps and agents, provisioning those primitives matters more than ever. neon.ts is our branch config and infrastructure-as-code file: declare…
You (And Your Agent) can provision Neon in Vercel CLI
The Vercel CLI now lets agents provision a database for your project with a single command from your terminal: 𝚟𝚎𝚛𝚌𝚎𝚕 𝚒𝚗𝚝𝚎𝚐𝚛𝚊𝚝𝚒𝚘𝚗 𝚊𝚍𝚍 𝚗𝚎𝚘𝚗
Shipping psql without psql: a pure-TypeScript Postgres client in neonctl
neonctl psql used to fail with "command not found" whenever psql wasn't on your PATH — which is most macOS laptops, slim Linux containers, Windows boxes, CI runners, and sandboxes. So we did the unco…
A branch-first dev loop for Neon: link, checkout and env pull
We're hard at work on the private preview of Neon Platform and along the way we're rethinking Neon's branch-first developer experience. First up: three new Neon CLI commands that turn branch-first de…
I Slop Forked Neon. You Should Too.
Agents are much better at using APIs, CLIs, and MCP servers than clicking through dashboards. That's why APIs matter even more in the age of agents. At Neon, we aim to expose all platform functionali…
We're including 5x more data transfer in all paid plans
Starting June 1, 2026, every Neon paid plan includes 500 GB of data transfer per month, up from 100 GB — enough to remove egress charges for most workloads. The change is automatic and appears on you…
May 2026
We're building the boring backend for apps and agents
Recently, everyone has been talking about throwing it all away and building entirely new magic sci-fi cloud infrastructure for agents. Building real infrastructure is hard enough as it is. AI has onl…
How Neon's lakebase architecture stays resilient to cloud failures
In the last year, agents have strained the limits of cloud infrastructure with new usage patterns: higher throughput of control-plane operations, more demand for on-demand infrastructure, and capacit…
Neon for Agent Platforms: A Companion Skill for the AI Agent Program
We've been working on neon-for-agent-platforms, a companion skill that complements our existing neon-postgres skill. The existing skill covers auth, drivers, branching, the Data API, MCP, and core Po…
Everyone gets faster writes: We turned off FPW's in Neon
We've managed to give customers up to 5x performance increase on write-heavy workloads by disabling full-page writes, a Postgres durability safety feature that is made redundant by Neon's own storage…
April 2026
Introducing Organization Spending Limits
Your team ships a new feature, traffic spikes, and autoscaling does its job. Great — until the bill arrives and it’s three times what anyone expected. By then it’s too late to do anything about it. M…
Agents grew up, so did our docs
A year ago, if you asked an agent about Neon, you got whatever it half-remembered from training. Now it goes looking and reads what it finds. Our docs were written for humans who scroll, not machines…
Neon is now available as an OpenAI Codex Plugin
An official Neon plugin is now available in the OpenAI Codex marketplace. It connects Codex directly to your Neon databases through MCP, so you can provision and manage Postgres databases without lea…
March 2026
Zero-Downtime Patching Part 1: Prewarming
Ensuring customer databases are always available is one of the most important things we do in Neon and Lakebase. We’ve designed the system with redundancy at every level, automatically failing over a…
Neon works with Stripe Projects for agentic provisioning
For most of 2025, AI coding agents got good at a specific thing: writing code. Give an agent a prompt, and it could scaffold an app, wire up an API, write migrations. But when the code was done, the…
How Specific Provisions Thousands of Databases for Coding Agents Using Neon
“I’m genuinely surprised by how well it handles that scale. You can create tons of databases and they’re available immediately. You can branch out immediately. All of those things make it really nice…
Ctrl-C in psql gives me the heebie-jeebies
There are a few different reasons to hit the brakes on a Postgres query. Maybe it’s taking too long to finish. Maybe you realised you forgot to create an index that will make it orders of magnitude q…
February 2026
15,000+ Apps Built Over WhatsApp: Inside QwikBuild’s Neon-Powered Architecture
“The biggest strength of Neon is how it decouples storage and compute and makes them independently scalable. When an app isn’t being used, the compute node can be put in idle mode at extremely low co…
Where Agents Meet Infrastructure: Encore, Leap, and Neon
From the start, the team at Encore has been focused on solving a simple problem: shipping production infrastructure shouldn’t require a dedicated platform engineering team. They set out to make deplo…
Building a Deep Research Agent with Neon and Durable Endpoints
Every AI lab is shipping research agents. OpenAI’s Deep Research, Perplexity, and Gemini’s research mode. These products are not simple RAG pipelines. Recent papers like DeepResearcher and Step-DeepR…
Neon Is a Cursor Plugin
Cursor just launched plugins, making it easier than ever to give Cursor structured access to external tools and infrastructure. Neon is part of the initial launch set: you can install the Neon plugin…
Neon Compute Autoscaling Report
View the Autoscaling Report here.
add-mcp: Install MCP Servers Across Coding Agents and Editors
A few weeks ago, Vercel released add-skill (now npx skills), a CLI for installing agent skills across different coding agents and editors like Claude Code, Cursor, and VS Code. It solves a very real…
How Zite Provisions Isolated Postgres Databases for Every User
“We were getting ready to hire dedicated engineers just to manage and scale Zite Database. With Neon, we didn’t need to do that – we were able to give every end user their own database, including on…