These benchmarks are used to evaluate the performance of SWC on every commit, as well as against other libraries like Babel or esbuild. You can view the benchmark source code here.Loading...
This is the most important benchmark, since build tool authors will run tasks concurrently. Note the performance of Babel and
tsx is indentical for parallel and synchronous benchmarks.
Not as critical as the parallel benchmark.
This benchmark suite does not use all CPU cores. For SWC, the benchmark stores only one task on the Node.js thread pool at a time.