Announcement of VoidZero – the next generation of JavaScript tools

Today was announced the next generation of JavaScript toolkit from the creator of Vue and Vite. The seed funding raised $4.6 million, with Accel and Amplify being the main investors. In light of this announcement, it is important to understand the context in which it occurs and the significance of the changes it brings.

Fifteen years ago, JavaScript was primarily a scripting language for browsers. Today, it has become the most widely used language used to create web applications, mobile applications, games, and even the Internet of Things (IoT). Despite the development of many tools to solve increasingly complex problems, the JavaScript ecosystem remains fragmented, making getting dependencies to work together one of the most challenging tasks.

Evan Yu invested significant effort into learning the JavaScript tool stack and developing abstractions to make things easier for developers, leading to the creation of Vite in 2020. In four years, Vite has become one of the leading web development tools, supported by multiple meta frameworks and boasting over 15 million downloads per week. Vite has established itself as a common infrastructure layer for the next generation of web frameworks.

The community's trust in Vite prompted the author to think deeply about the future of the project. Despite significant improvements in the developer experience, Vite still relies on a variety of abstractions and workarounds to iron out inconsistencies between dependencies. Internally, it suffers from duplicate parsing and serialization, which limits its performance. Attempts to adapt the new Rolldown bundler to Vite's needs showed that Vite's problems reflect general problems of the JavaScript ecosystem: fragmentation, incompatibility and inefficiency. To radically change the situation, a unified toolkit is needed.

Now imagine this set of tools:

• Unification: Use the same AST, resolver and module interop for all tasks (parsing, transforming, linting, formatting, packaging, minification, testing), eliminating inconsistencies and reducing unnecessary parsing costs.

• High performance: written in a compiled language, designed from the ground up for speed, maximum parallelization, and support for low-cost JS plugins. The money spent on productivity unlocks more ambitious capabilities that improve not only the developer experience, but also the end user experience.

• Composability: Each toolchain component can be used independently of each other, offering building blocks for advanced customization.

• Runtime independent: Not tied to any specific JavaScript runtime—provides the same experience for developers in all environments.

This set of tools will not only improve Vite, but will also lead to significant improvements across the entire JavaScript ecosystem. This is an ambitious vision and requires a full-time team to realize it, something that was not possible within the independent sustainability model of my past projects. This is why VoidZero was founded.

Over the past year, a team has been assembled with deep experience in JavaScript tooling, including the creators and core contributors to such widely used open source projects as Vite, Vitest, Oxc as well as former core members Rspack .

Hard work is underway to create the fundamental elements of the planned tools. In addition to ongoing improvements to Vite, the following milestones have been made:

• The fastest and most spec-compliant JavaScript parser (oxc-parser), 3 times faster than SWC – benchmark

• Fastest Node.js compatible resolver (oxc-resolver), 28 times faster than enhanced-resolve – benchmark

• Fastest TypeScript/JSX converter (oxc-transform), 4 times faster than SWC – benchmark

• The fastest linter (oxlint), 50-100 times faster than ESLint – benchmark

• The most powerful tool for running web application tests (Vitest)

• The fastest collector (Rolldown), built on the basis of Oxc. Faster than esbuild and all other Rust bundlers – benchmark (currently in alpha)

These projects are used by leading engineering teams from companies such as OpenAI (ChatGPT web client), Google, Apple, Microsoft, Visa, Shopify, Cloudflare, Atlassian, Reddit, HuggingFace, Linear and many more.

Further development plan

The main goal for the coming months is to stabilize Rolldown and make it a single bundler for Vite in both development and production. Significant progress has already been made, and the plan is to release an alpha version of Vite with Rolldown support by the end of this year.

In 2025, work will continue on other planned Oxc features (minification, formatting), as well as the gradual transition of the entire Vite ecosystem to Rolldown and Oxc. There will be close collaboration with ecosystem partners and stakeholders to ensure a smooth transition for end users.

Everything published publicly will remain available. In addition to open source projects, it will offer a comprehensive JavaScript tooling solution specifically designed to meet the scale and security requirements of enterprise systems.

Answers to basic questions:​

• What is the connection between these open source projects and VoidZero?
  The management of Vite and Vitest remains the same as before. Both core teams include members working in various organizations (VoidZero, StackBlitz, NuxtLabs, Astro). VoidZero Inc. recruits several core members of Vite and Vitest.

• What about Vue?
  VoidZero as a business is completely separate from Vue. Vue will remain an independent project, but will receive priority support from the new toolkit developed by VoidZero

• Why Oxc and not SWC?
  Many of our team members have made extraordinary contributions to SWC in the past. Beyond the pure performance advantage, Oxc has a number of fundamental design differences from SWC that make it a better foundation for the end-to-end tooling we're building. We will share more technical information on this topic in future blog posts. Follow the news!

• Why Rolldown instead of esbuild/Rollup?
  We needed a bundler that was very fast, good for building applications, and fully compatible with the Vite plugin ecosystem. This is discussed in detail in the Rolldown documentation. Building Rolldown on top of Oxc also opens up the ability to perform more AST-related tasks in parallel during the batching phase, such as emulating and batching dts with isolatedDeclaration: true.

• How will this be different from previous attempts to create a unified JS toolkit?
  Unified Toolkit's biggest problem is the “ground zero” problem: it needs to reach critical mass for exponential adoption to justify further development, but it's hard to cross the chasm before it actually realizes its vision. VoidZero doesn't have this problem because Vite is already the fastest growing toolkit in the JavaScript ecosystem.