Postgres is the world's most popular open source relational database. Traditionally it runs on the server and scales out using read replicas.

Thanks to PGlite, our new lightweight WASM build of Postgres, Postgres now runs efficiently, with persistence, in the web browser and other client environments like Node, Bun and Deno. As of v0.11 just released today, ElectricSQL now supports syncing data between Postgres in the cloud and PGlite in the client.

This means that you can now build apps that have a partial replica of a Postgres database in process, with zero network latency, realtime reactivity and background sync. It means you can run Postgres in process, inside your scripts, your development environments, your edge workers and your web services. Just using npm install.

We've just published version 0.10 of Electric. This is the first release that properly supports where-clause and include-tree filtering with Shape-based sync.

It also adds data type support for byte arrays and blobs.

To set the tone for 2024, we're kicking January off with a fresh release of Electric that introduces a whole slew of improvements to the experience of configuring, deploying, and developing with Electric.

This new version addresses an array of issues that have been reported by our community and early adopters. Thanks to everyone who has provided feedback and bug reports!

Version 0.8 of ElectricSQL has now been released. This release improves data model and deployment compatibility, including support for JSON and running on popular managed Postgres and application hosts, such as Supabase and Digital Ocean.

Here at Electric, we are thrilled to have published version 0.7 of ElectricSQL, which includes the new migration proxy and extended type support. This release lays the groundwork for many of the new features we are currently working on, and which will be available in future versions.

Introducing the v0.6 release of ElectricSQL. A local-first sync layer that you can use to build reactive, realtime, offline-capable apps directly on Postgres.