Bun's Rewrite from Zig to Rust Passes 99.8% of Tests — What's the Other 0.2%?
Bun's team decided Zig was too niche and rewrote the runtime in Rust — tests pass almost entirely, but that 0.2% is probably a closure at 3 AM.
News that will make Zig fans twitch and Rustaceans grin: Jarred Sumner, Bun's creator, announced that rewriting the runtime from Zig to Rust now passes 99.8% of the test suite. Yes, you heard that right — another cult project swaps its language under the hood, like a teenager changing their Steam avatar.
In short: Bun is a fast JavaScript runtime originally written in Zig to outrun Node.js and Deno. But apparently, the Zig ecosystem felt too "hard to hire for" (familiar pain, right?). Rust, on the other hand, is like IKEA: plenty of ready-made modules, but you still can't assemble a chair without instructions. 99.8% of tests pass, meaning the remaining 0.2% are probably some edge case with closures at 3 AM.
What does this mean for developers? If you write JS/TS, Bun might get even faster and more stable soon. If you're a Zig fan — hang tight, maybe your favorite language will stay only in hardcore embedded projects and enthusiasts' hearts. Either way, migrating to Rust is like moving to a new apartment: boxes of code are moved, but somewhere you lost that mug with the Zig logo.
METABYTE studio comment: Rewriting a runtime is like refactoring legacy code: painful at first, profitable later. If your project needs a tech stack change, we know how not to lose that 0.2% of tests along the way.
NEXT STEP
Liked the approach?
We apply the same principles to client projects: AI, automation, products that don't die after launch.