Skip to main content
Use WebAssembly in your extension without custom bundler setup. Extension.js ships with built-in WebAssembly (Wasm) defaults in its Rspack pipeline, so common Wasm workflows work without extra configuration.

​
When WebAssembly is a good fit

  • You run compute-heavy tasks in your extension (parsing, media, optical character recognition (OCR), transforms).
  • You need near-native performance for frequently executed code paths.
  • You are reusing existing Wasm modules from web projects.

​
WebAssembly capabilities

​
Template examples

​
transformers-js

transformers-js template screenshot AI/ML sidebar extension using Transformers.js with WebAssembly-powered inference.
Repository: extension-js/examples/transformers-js

​
What Extension.js enables

The Wasm plugin configures:
  • experiments.asyncWebAssembly = true
  • .wasm in module resolution extensions
  • Path aliases for common Wasm libraries (for example, ffmpeg for media processing, imagemagick for image manipulation, and tesseract for OCR) so imports resolve correctly at runtime.
This reduces boilerplate when importing Wasm modules from extension scripts/pages.

​
How to use

Use Wasm modules in your extension code as part of regular development/build flows:
You can also test against a known public Wasm sample:

​
Typical use cases

  • Compute-heavy parsing/transforms.
  • Media processing pipelines.
  • OCR / image processing workflows.
  • Performance-sensitive algorithms shared across web/runtime contexts.

​
Best practices

  • Keep Wasm modules focused on hot paths; keep orchestration logic in JavaScript/TypeScript.
  • Validate output size and load timing for extension contexts (background, content scripts, pages).
  • Test Wasm-heavy flows across browser targets you ship (chrome, edge, firefox).

​
Next steps