With BUN 1.2, we're introducing new features and APIs that make it easier than ever.

We've also spent a lot of time improving stability and Node.js compatibility.

That's why companies like X are using BUN in production.

BUN lets them ship features at scale in days instead of months.

Before we talk about the new features in BUN 1.2, let's start with an update on Node compatibility.

BUN is designed to be a drop-in replacement for Node.js.

BUN supports all the major Node.js APIs like FS, Net, and HTTP.

In BUN 1.2, we've added support for even more APIs, including DGRAM, HTTP2 Server, and Cluster.

We've also made some major performance improvements.

In BUN 1.2, a Hello World Express server handles three times more requests per second than in Node.js.

In BUN 1.2, we started to run the entire Node.js test suite for every change we make to BUN.

It's a suite that includes thousands of tests from Node's repository and ensures that if code works in Node, it also works in BUN.

For each of the following Node.js modules, BUN passes over 90% of Node's test suite.

Every day, BUN gets closer to 100% Node compatibility, and we're excited to share more progress on this very soon.

BUN is designed as a cloud-first JavaScript runtime.

That means providing built-in APIs for all the tools you need when running a production application in the cloud.

That's why we're excited to introduce S3 object support in BUN.

S3 is the de facto standard for object storage and even predates Node.js.

Think of it as the file system API for the cloud.

There are dozens of S3-compatible services that you can use in production or even on your local machine.

In BUN 1.2, you can use the new BUN S3 API to read and write files from an S3 bucket.

It uses the same simple APIs as BUN file.

It also extends the web standard APIs.

That means you can pass it to any API that accepts blobs, like requests and response.

We've also made it possible to read and write S3 files with just a URL.

BUN looks for well-known environment variables to configure your credentials and bucket.

If you want to customize and create your own S3 client, you can import the S3 client API.

BUN's S3 client is implemented in highly optimized native code.

The S3 client in BUN can download files in parallel up to five times faster than the leading MPM package can on Node.js.

Just like object storage, another data store that's widely used in production is Postgres.

And that's why we're excited to introduce built-in Postgres support for BUN.

In BUN 1.2, we're expanding the BUN SQLite APIs by introducing BUN SQL.

BUN SQL is a fast built-in SQL client for BUN.

Postgres is the first database we support, and we have a PR to add MySQL very soon.

BUN SQL's APIs use tagged template literals to create queries.

This prevents SQL injection by design without having to manually escape strings.

BUN SQL's APIs are inspired by the popular MPM package, Postgres.js.

If you're already using Postgres.js, you can switch to BUN SQL by changing the import.

Just like BUN S3, BUN SQL is implemented in native code instead of JavaScript.

It can read rows up to 50% faster than the leading MPM package can on Node.js.

BUN isn't just a JavaScript runtime.

It's also an MPM compatible package manager.

BUN install is designed to work with Node.js and BUN.

Even if you aren't ready to migrate your application to BUN, you can still use BUN install with Node.js.

In BUN 1.2, we're introducing the biggest update yet to BUN install, a text-based log file.

When BUN install was first designed, we decided to use a binary log file for performance reasons, but we heard a lot of your feedback that they're impossible to review in pull requests, hard to fix merge conflicts, and difficult for tooling to use.

That's why BUN's new text log file is a JSON C file, which is JSON with support for comments and trailing commas.

This makes it much easier to view diffs in pull requests, and trailing commas make it much less likely to cause merge conflicts.

Some projects start out being faster, then get slower as they add new features.

BUN is not one of those projects.

That's why we spent a lot of time figuring out how to make the text log file just as fast.

In BUN 1.2, we made cached installs 30% faster.

For new projects, BUN install will default to generating a text log file.

For existing projects, BUN will continue to use the binary log file.

To migrate to the text log file, you can pass the save text log file flag to BUN install.

BUN is a complete toolkit for building full stack applications with JavaScript and TypeScript.

And front-end is a really important part of your stack.

In BUN 1.2, we're introducing HTML imports.

HTML imports simplify your entire front-end tool chain into a single import statement.

BUN scans the HTML and automatically minifies and bundles JavaScript and CSS.

Modern tools like React, TypeScript and Tailwind just work.

Together with HTML imports, S3 and BUN SQL, building full stack applications with JavaScript has never been simpler.

We're really excited to see what you build with BUN.

There's so many more features and improvements in BUN 1.2 that we can't go over them all.

Check out our blog for everything else that we shipped in BUN 1.2.

To get started with BUN, run any of the following commands in your terminal.

That's BUN 1.2.

And for BUN, it's still just the beginning.