Ladybird Is Building the First New Browser Engine in Years

Ladybird is the first serious attempt in over a decade to build a web browser on a genuinely new engine — one that shares no code with Google's Blink, Apple's WebKit or Mozilla's Gecko. Built from scratch by a nonprofit and paid for entirely by donations, it is targeting a developer alpha in 2026 on Linux and macOS. It will not challenge Chrome on day one. What it challenges is the quieter assumption that a new browser engine cannot be built at all anymore.

• Ladybird is a from-scratch engine — its own LibWeb renderer and LibJS JavaScript engine — not a fork of Chromium, WebKit or Gecko.
• It is run by the nonprofit Ladybird Browser Initiative and funded by unrestricted donations (Cloudflare, FUTO, Shopify, 37signals), with no search deals, ads or crypto tokens.
• The benchmarks are real: it crossed 90% of the Web Platform Tests in October 2025 and passes roughly 2.07M WPT subtests as of April 2026 — about a third of what Chrome and Firefox pass.
• Roadmap: developer alpha 2026 (Linux, macOS), beta 2027, stable 2028 — with a Rust migration and a contentious closed-contribution model along the way.

Why the web needs a fourth engine

The browser is the most important application in computing, and almost every one is built on an engine controlled by a company with a commercial agenda. Chrome, Edge, Brave, Opera and Arc all run on Blink. Every browser on iOS runs on WebKit. Firefox and Tor run on Gecko. That means the practical rules of the web — what ships, what is prioritized, what quietly gets deprecated — are set by three firms whose incentives do not always match users'. When Google can shape a standard through Blink's sheer reach, there is no independent implementer positioned to say no. Ladybird matters because an engine with no ad business, no search default and no platform to defend can make different calls, and its mere existence is a check that a brand-level browser choice can never be.

RelatedFirefox Halves Its Release Cycle to Two Weeks

Servo is the honest comparison point — the last from-scratch engine to get real attention — and the contrast is instructive. Servo, now under the Linux Foundation, is a superb Rust rendering engine but not a full browser you can switch to; it is meant to be embedded. Ladybird is trying to do the harder, less glamorous thing: ship a complete, everyday browser, not just an engine demo.

What makes Ladybird different under the hood

Ladybird grew out of SerenityOS, a from-scratch hobby operating system, and its architecture is deliberately independent. LibWeb is a brand-new rendering engine and LibJS is a brand-new JavaScript engine with its own parser, bytecode compiler and interpreter. It uses a security-first multi-process design — separate sandboxed processes for web content, networking, image decoding and compositing — so a crash or exploit in one renderer does not take down the browser or reach the rest of your machine. The codebase began in modern C++, and in February 2026 the team shipped a Rust reimplementation of the LibJS frontend — the lexer, parser, AST, scope collector and bytecode generator — enabled by default, the first major payoff of an ongoing migration to a memory-safe language for the most security-sensitive code a browser runs.

None of that would matter if the engine could not render the real web, which is why the test numbers are the story. In October 2025 Ladybird crossed 90% of the Web Platform Test subtests for the first time — the rough readiness bar Apple has used internally — and by April 2026 it was passing about 2.07 million subtests, alongside 97.8% of the imported test262 JavaScript-conformance suite. That is roughly a third of the six-to-seven million Chrome and Firefox pass: a real gap, but a remarkable one for a donation-funded project only now reaching alpha.

The road from a hobby OS to an alpha

1. 2019SerenityOS begins. Andreas Kling starts the from-scratch OS whose browser becomes Ladybird.
2. 2024Spun out as a nonprofit. The Ladybird Browser Initiative forms to build a standalone, independent browser.
3. Oct 2025Crosses 90% of the Web Platform Tests. Passes the threshold Apple has used as a rough readiness proxy.
4. Feb 2026LibJS frontend reimplemented in Rust. Lexer, parser, AST, scope collector and bytecode generator, on by default.
5. Jun 2026Closes public pull requests. Maintainer-only development, citing AI-generated contributions and browser security.
6. 2026Developer alpha — Linux & macOS. First shippable build, aimed at developers and early adopters.
7. 2027Beta.
8. 2028Targeted public stable release. Windows and mobile support planned to follow.

Can Ladybird realistically compete with Chromium?

Not on Chromium's terms, and it is not trying to. Google pours thousands of engineers and effectively unlimited search revenue into Blink; Ladybird has a small team and a donation jar. Judged as a Chrome-killer it will lose, and its own founder is careful never to make that claim. Judged by its actual goal — proving a modern engine can still be built independently, and giving the web one implementer with no ads or search deal to protect — it is already succeeding. The most interesting move of 2026 underlines how seriously it takes that independence. On June 5, 2026, the project stopped accepting public pull requests from non-maintainers. Founder Andreas Kling framed it as a reluctant response to AI tooling flooding open source with contributions that look serious but are not, eroding the effort-equals-good-faith signal maintainers once relied on — a real security concern for software that runs untrusted code from the whole internet. It trades openness for control at a delicate moment, and it is the clearest sign yet that Ladybird will guard its engineering culture even when that is unpopular.

RelatedThe Anti-AI Browser Backlash Is Fueling DuckDuckGo's Surge

What to watch before the 2026 alpha

The next two to three years will decide whether Ladybird is a landmark or a footnote. Five questions matter more than the rest.

Our take

Ladybird is the most important browser project almost nobody is using yet, and that is precisely why it deserves attention. The engine monoculture is a genuine structural problem, and a from-scratch engine funded without ads or search deals is the only kind of answer that addresses the root cause instead of reskinning someone else's renderer and calling it an alternative. The ~2.07M WPT subtests prove this is engineering, not vaporware; the refusal to monetize users is a stance the incumbents structurally cannot match. The risks are honest ones — timelines slip, donations are harder than search revenue, and the closed-contribution turn is a bet. But the web needs this to exist, and Ladybird is the closest anyone has come in years. A 2026 alpha will not dethrone Chrome. It will prove a different kind of browser is still possible, and right now that proof is worth a great deal.

Original analysis by GenZTech. Figures current as of July 2026.