Effect 4.0 Ships: Dependency-Free Rewrite with 5x Smaller Bundles

Effectful Technologies released Effect 4.0 on 30 September 2026, the first stable version of its TypeScript framework for production applications with typed errors, structured concurrency and dependency injection. The line spent eight months in beta and six weeks as a release candidate before that date. Daniel Curtis, in InfoQ's report of 10 November 2026, describes a from-scratch rewrite of the fiber runtime aimed at smaller bundles, lower memory use and higher throughput.

This piece walks through the 4.0 size and speed figures, the consolidation of @effect/platform, @effect/rpc and @effect/cluster into the main effect package, and the support rules set against September 2029 and Effect 5.0. It then covers the launch download counts, a Hacker News note from the beta, the 8 October stability-tag fixes in 4.0.1 and 4.0.2, and Effect 4.1 on 20 October 2026. It is for frontend and backend teams on Effect 3.x, and for developers comparing that stack with ZIO or with Effection 4.0.

01Five times smaller bundles, 6.4x throughput

Effect 4.0 replaces the fiber runtime from scratch. A minimal program using Effect, Stream and Schema is 7.1 kB, against 35.6 kB in 3.x, a 5x reduction on the team's figures. The heap used by 50,000 fibers drops from 157.5 MB to 21.8 MB, an 86% cut.

Concurrent task throughput moves from 0.71 million tasks per second to 4.57 million tasks per second, roughly 6.4x. Those three results are medians of nine fresh-process runs on identical source, published as the team's own figures. Stork.ai notes that the bundle, heap and throughput figures have not been independently reproduced.

02Platform, RPC and cluster inside effect

Alongside the rewrite, the ecosystem has been consolidated. Core modules from @effect/platform, @effect/rpc and @effect/cluster now live inside the main effect package, and every @effect/* package shares a single version. The Effect core that now holds those modules ships with zero runtime dependencies.

That followed the final removal of pg, replaced by a native PgProtocol client, and the removal of MessagePack from the runtime. The native PgProtocol client is what replaces pg. MessagePack is no longer a runtime dependency of the Effect core.

03September 2029 and the 5.0 dates

With 4.0, Effectful Technologies introduced a long-term support policy for the 4.x line. Effect 4.x receives bug fixes until September 2029, or one year after 5.0 ships, whichever is later. If the one-year mark after 5.0 is earlier than September 2029, bug fixes still run until September 2029; if it is later, they run until that later date.

Security fixes run until September 2029 or two years after 5.0 ships, again whichever is later. The two-year mark after 5.0 is compared with September 2029, and security fixes continue until the later of those dates. Effect 4.0 is the first stable release of the 4.x line covered by this policy.

04First-day downloads and a beta comment

Stork.ai points out that the 43.9 million weekly downloads cited at launch cover the whole Effect ecosystem across beta and RC tags. Stable Effect 4.0 itself recorded "approximately 150,000 downloads on its first day", the first-day figure Stork.ai gives for the stable release.

A Hacker News commenter who has been on the beta wrote, "I've been using the beta on three separate projects for quite a while now and it has been totally pain-free." They mostly use the v3x core library, which has expanded to contain other libraries in v4, and Schema. They said they cannot speak to much outside those areas, though they trust the team has done a great job all around.

The same person said 4 "comes with a much simpler API in some modules" and that "Schema is much better," calling the release an exceptional update. They followed many of the video logs on YouTube, described the rebuild as incredibly well done, and named the project one of their favourites. They hoped a lot of people would try it and adopt it, especially with the LTS.

05First-week tags, Effect 4.1, migration

The team's post, Effect 4.0: The First Week, was published alongside patch releases 4.0.1 and 4.0.2 on 8 October. It admitted that several modules intended to ship as unstable went out without a @stability tag, implicitly promising they were stable. The missing tags were added, every exported API now declares its stability explicitly, and HTTP, SQL, RPC and AI spans moved to the OpenTelemetry semantic conventions.

Effect 4.1 followed on 20 October 2026 with new effect/crypto modules covering randomness, hashing and signing. The same release added a Glob module backed by FileSystem.glob on Node, Bun and Deno, plus a streaming Csv module in effect/encoding. HttpApi gained support for RFC 9457 Problem Details.

Teams on 3.x start from the v3 to v4 migration guide, and the team suggests coding agents can handle most of the mechanical work. Effect sits in similar territory to ZIO in the Scala world, alongside TypeScript libraries such as Effection 4.0 that target structured concurrency with a deliberately smaller surface. Effectful Technologies maintains the open-source framework, including observability primitives, for both frontend and backend applications.