Hey, it’s been almost a couple of months since the last update and for a good reason: EdgeDB 6 is shaping up to be an incredible release, and we’re busy making sure it will rock.

• Dijana is updating the built-in ai extension with the latest models and function calling. Her improvements to our “Ask AI” feature are set to ship soon!

• Andrey, our new hire, is adding a new extension aptly called vectorstore to make it a one-minute task to use EdgeDB as a vector database. He also created adapters for LangChain and LlamaIndex, which we’re testing now internally.

• Scott, after implementing the new net::http module and webhooks support for Auth, is maintaining momentum by enhancing security for Auth API endpoints.

• Scott, Sully, and Zack are working on separate components to add built-in SMTP support for our auth extension for EdgeDB Cloud. This will simplify getting started with email-based authentication in the cloud.

• Matt is living the every Rust’s engineer dream – rewriting existing perfectly working software to Rust. Jokes aside, EdgeDB 6 will benefit from a new Rust-based HTTP layer, and there’s also a lot of work done towards rewriting EdgeDB 7 IO completely in Rust. We’re crossing our fingers!

• Speaking of dreams, Sully just realized his by landing an implementation of his RFC to simplify Path Resolution in EdgeQL. Although this can be considered a minor breaking change, it will allow us to evolve EdgeQL much quicker.

• Aljaz and Victor are finalizing the 6.0 mega project: adding full SQL support. They’re thrilled to spend their days exploring different popular ORMs and discovering how we can support them in the best way possible.

• Aleksandra has recently created a new NextJS+EdgeDB template that should soon be available on vercel.com. She’s now busy preparing us for React Summit in NYC on November 18th. Excited to see some of you there!

• Fantix, just before he left on 🏝️, finalized his work on a new slow queries log feature.

• Zack and the cloud team are working on our cloud infra non-stop, making sure it operates smoothly, while preparing to ship pretty significant internal improvements.

• Daniel is about to land a new major functionality to EdgeQL: functions that can have insert / update / delete commands in them.

• James is almost done with a huge round of updates in our UI. Soon you’ll be able to show/hide, reorder, and pin columns in Data View. He’s now busy with creating a UI for slow queries log.

• Roman is designing several semi-secret projects, including a page for the upcoming Python hackathon (this is a secret!)

Welcome to EdgeDB Dev Report – bi-weekly or so updates about the latest in EdgeDB development. Here’s some bites for you:

• Scott is working on speccing and implementing a new std::net:: module to bring HTTP requests and 🪝 Webhooks to EdgeDB.

• Dijana is about to complete a “mini-StackOverflow” section of our Documentation, fed by questions you ask on our 💬 Discord.

• Matt is rewriting our EdgeDB <> PostgreSQL communication in 🦀 Rust. Just recently he RIIR’ed our backend connection pool and we’re about to make the new pool the default.

• Aleksandra is creating a swath of new project templates for Next.js and EdgeDB. From time to time we find her reading the Replicache docs for no apparent reason 😉.

• Sully is putting last finishing touches on the new “in-place upgrades” feature of EdgeDB 6. So you can update your database server version without running full dump/restore (for some reason our users want that 🤯.)

• Aljaz is close to finalizing his epic project on implementing write SQL support in EdgeDB. He’s currently busy enabling 🔐 access policies for SQL queries.

• Roman has started designing a website for a new secret project. Completely out of the blue he’s researching molecular properties of gels and liquid metals 🔬.

• Zack, Elvis, and the cloud team are keeping our Cloud running and working on improving our infra. BTW, we’re looking for experienced SREs and Cloud Engineers 👷‍♀️, drop us a line.

• James was last seen adding major new UI features 🫡 for EdgeDB and the Cloud.

• Victor is wrapping up his work on bringing 📍 PostGIS to EdgeDB 6.

We’ve recently added a bunch of updates to EdgeDB Cloud, including these few highlights:

• A new branch graph that lets you visualise your database branches, just like Git, and allows you to easily navigate the migration history of any branch. If you have the GitHub integration installed, we now also show you which of your database branches are automatically mirroring your branches in GitHub, showing you their sync status, and letting you jump straight to the associated PR.

• An updated billing page with an improved upcoming invoice section, which now gives a fine-grained breakdown of your current and estimated usage and costs for the month.

• Don’t like the name of your cloud instance? You can now rename instances from the instance dashboard.

• The secret keys page has received a bit of a redesign to make it easier for you to manage your personal secret keys, logged in web and CLI sessions, and secret keys being used by installed integrations.

There have also been many other small bug fixes, and we’ve given some pages a fresh coat of paint to make everything a little more consistent.

Today we’re excited to announce an EdgeDB Cloud + Vercel Integration and our first ever Hackathon! Read more here.

In EdgeDB 5, EdgeDB Auth gains a couple of passwordless providers — WebAuthn and Magic Links — as well as some new OAuth providers — Slack and Discord. Read more here!

In EdgeDB 5, the new ext::ai extension automates embeddings generation and has native integration with OpenAI, Anthropic, and Mistral. Read more here!

With EdgeDB v5, we’re rethinking the concept of a “database” within a database server (goodbye ’70s!) to a “branch.” This change not only simplifies our vocabulary but also shifts our mindset, opening up a host of possibilities for the tools we can now build. Read more in the announcement blog post!

We’re thrilled to announce the general availability of the EdgeDB Cloud Free Tier. This marks a significant step forward in making EdgeDB more accessible and competitive. But what really sets EdgeDB Cloud apart? Read the announcement blog post for details and a bit of benchmarking!

Many parts of the EdgeDB workflow involve using our CLI, and installing the CLI is the very first step in starting to use EdgeDB. Today, we’re making it easier than ever to install and invoke the EdgeDB CLI for JavaScript projects.

We now provide a wrapper script as a bin entry in our JavaScript package published to npm. This wrapper will:

1. Install the CLI if it’s not already installed

2. Invoke the CLI with the arguments you pass to it

This approach eliminates the need to manually set up your PATH after installation and can simplify CI/CD workflows.

Here’s how to use it:

$ 

npx edgedb project init

# or with yarn

$ 

yarn edgedb project init

# or with pnpm

$ 

pnpm exec edgedb project init

Additionally, if you have scripts defined in your package.json, any invocation of edgedb will be forwarded to the wrapper script, which will install and/or call the CLI for you.

For example, you might have defined a migrate run-script that runs the code generator after migrating your database:

{
  "scripts": {
    "generate:querybuilder": "generate edgeql-js",
    "generate:interfaces": "generate interfaces",
    "generate:queries": "generate queries",
    "generate:all": "$npm_execpath generate:querybuilder && $npm_execpath generate:interfaces && $npm_execpath generate:queries",
    "migrate": "edgedb migrate && $npm_execpath generate:all"
  }
}

Running this migrate run-script with your package manager’s script runner will now invoke this wrapper script.

We’re launching the EdgeDB Cloud EU region.

Today we are announcing EdgeDB 4.0 and launching our Cloud product.

We’re kicking off this launch with a keynote presentation. Alongside discussing EdgeDB 4.0 and the Cloud, we’ve invited four amazing engineers and founders to share their stories about building real-world products with EdgeDB.

Read more in the announcement blog post!

We’re thrilled to announce EdgeDB 3.0, a release that comes loaded with new features and improvements:

• A simplification of EdgeDB Schema language.

• Triggers and mutation rewrites.

• UI gains a new Web REPL and a visual query builder.

• EdgeQL gets a new analyze command, splats, nested modules, new set operators, and more.

• edgedb watch command to simplify schema prototyping.

• ext::pgvector for embeddings.

• SQL in read-only mode.

• Native Java and Elixir clients.

Read more in the announcement blog post!

We are excited to announce that EdgeDB has raised a $15m Series A financing round! Continue reading to learn what this means for the product and the team behind it.

Today, just under 6 months after the release of 1.0 🏁, we’re excited to announce EdgeDB 2.0.

This release brings a slew of new features, including:

• a built-in admin UI;

• a top-level GROUP statement in EdgeQL;

• object-level security;

• range types;

• an official Rust client library (at long last!)

• …and a lot more.

Read our announcement blog post for more details and the HN discussion.

We are open-sourcing a new benchmarking rig called IMDBench. The goal is to evaluate the performance of major ORM libraries with realistic queries that would be required for a hypothetical IMDB-style movie database application.

Read our announcement blog post.

Today, after several years of building (and a long list of prereleases) we are extremely proud to announce the release of EdgeDB 1.0, the first open source, graph-relational database! 🎊

Read our announcement blog post and the HN discussion.