Effect 4.0 lands with 5x smaller bundles, 6.4x throughput and zero runtime dependencies

Effect 4.0 rebuilds the core for speed and size

Effect has shipped version 4.0 of its TypeScript ecosystem, a release the team describes as its most ambitious to date because the library was rebuilt from the ground up. According to the announcement on the Effect blog, which surfaced on the Hacker News front page, the rewrite targets three dimensions at once: bundle size, throughput and memory.

The published benchmarks compare bundles built from identical source, minified and gzipped, with runtime figures taken as medians of nine fresh-process runs:

• A minimal program drops from 35.6 kB under Effect 3.x to 7.1 kB under 4.0, roughly a fivefold reduction.
• Concurrent task throughput rises from 0.71 million to 4.57 million tasks per second, a 6.4x improvement.
• The heap required for 50,000 fibers falls from 157.5 MB to 21.8 MB, which the team frames as 86% less memory per fiber.

One versioned ecosystem with no runtime dependencies

The release also reorganizes how Effect is packaged. Many packages that previously shipped separately now live inside the core effect package, and the surrounding packages share a single version number and release in lockstep, which should reduce version-skew problems across the ecosystem.

The core package now has zero runtime dependencies. The team presents this partly as a supply-chain decision: because it controls every line it ships, there are no third-party dependency chains and less exposure to supply-chain attacks.

From single functions to clusters

Effect started with a much smaller scope, but the project now spans the range from a single function to a distributed system. The foundation is an effect system with built-in typed errors, dependency injection, resource management, structured concurrency and observability. A standard library and platform integrations grew around it, followed by higher-level distributed primitives such as durable workflows and clustering. The team's claim is that every layer builds on that same model, with the same guarantees all the way up.

Adoption and community growth

The blog cites 43.9 million weekly npm downloads for the week of September 21, 2026, growth of 179x since the 3.x line. It also reports that 4.x already accounts for 56% of downloads over the preceding seven days, against 44% for 3.x. Production use now includes companies of all sizes, including large enterprises, and separate community projects have formed around the ecosystem, among them Alchemy for cloud infrastructure and Foldkit for the frontend.

Long-term support becomes policy

Starting with this release, every major version of Effect carries a long-term support policy. For 4.x that means bug fixes until September 2029, or one year after 5.0 ships, whichever is later, and security fixes until September 2029, or two years after 5.0 ships, whichever is later. Support lasts at least three years and can extend further if 5.0 ships late. The team says the full policy and its release process will be detailed in a follow-up post.

What comes next

The first priority is stabilizing the ecosystem. Some newer modules are still tagged @stability unstable or @stability experimental, meaning they may change in minor or patch releases; the plan is to refine them with production feedback and promote them to stable. Beyond that, the roadmap calls for broader native platform support, more higher-level primitives and a single programming model for entire applications. For teams migrating from 3.x, the team points to its migration guide and suggests handing it to a coding agent, which it expects to handle most of the work on its own.

Why it matters

With tens of millions of weekly npm downloads, changes of this scale in Effect reach a large share of the TypeScript world, and the benchmark gains target the two objections most often raised against effect systems in production: runtime overhead and payload size. The zero-dependency core speaks to ongoing supply-chain anxieties in the npm ecosystem, while the long-term support policy is clearly aimed at enterprises that need multi-year planning horizons before betting a product on a library. The adoption split, with 44% of downloads still on 3.x, suggests a substantial migration wave is still ahead, and the team's bet that AI coding agents can automate most of that migration will now be tested in practice.