npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

rapier-jxl

v3.1.0

Published

JavaScript JPEG XL encoder for single-file HTML apps and photography tools. Lossless and lossy RGBA, exact alpha, HDR, and optional WebAssembly. Offline, no runtime dependencies.

Downloads

1,271

Readme

Rapier JXL

Rapier's complete JPEG XL encoding system in 50,021 gzip bytes: lossless and lossy encoding, native precision, HDR, exact alpha, JPEG transcoding, optional WASM and parallel workers. JavaScript, zero runtime dependencies, offline. MIT license.

This is the complete worker used by Rapier. The separate core is 11,670 gzip bytes. Both can be embedded in one HTML file.

npm install rapier-jxl

Builds

| Build / import | Gzip bytes | Capabilities | | --- | ---: | --- | | Rapier worker · rapier-jxl/rapier/worker | 50,021 | Core + effort 1–9, optional WASM, Photo, JPEG transcoding and parallel lossless groups. | | Rapier module · rapier-jxl/rapier/min | 50,075 | The same system with an encoder factory and worker installer. | | Core · rapier-jxl/min | 11,670 | Lossless or lossy typed RGBA, native precision, HDR and exact alpha. | | Effort · rapier-jxl/effort/min | 30,013 | Core with lossless compression search, efforts 1–9. | | WASM · rapier-jxl/wasm/min | 36,336 | Effort with inlined WASM and JavaScript fallback. | | Photo · rapier-jxl/photo/min | 15,837 | Lossy photographs from RGBA pixels. | | JPEG · rapier-jxl/jpeg/min | 13,882 | Existing JPEGs, preserving admitted coefficients and orientation. | | Metadata · rapier-jxl/metadata/min | 2,560 | Attach, replace or remove caller-supplied Exif/XMP. |

Each size measures a complete standalone file at gzip level 9. Readable imports such as rapier-jxl/rapier share dependencies when bundled together; separate minified files each include theirs. Optional JPEG and Photo ANS builds, combined sizes and file hashes are listed in Entry points and embedding.

Used by Rapier: the app embeds the exact published Rapier worker bytes. Pixel encoding defaults to effort 9 and quality 90; lossless: true selects quality 100. JPEG transcoding uses effort 9. WASM falls back to JavaScript; larger lossless images use available helper workers. Complete API.

Complete worker versus libjxl

Lossless RGBA8, including hidden RGB, through the complete published Rapier worker, compared with native libjxl 0.11.2. Effort numbers are each encoder's search budgets. Bpp means bits per pixel. Rapier ms is one call per configuration without warmup; libjxl ms is the median of three calls after one warmup.

| Input | Effort | Rapier bytes / bpp / ms (single) | libjxl bytes / bpp / ms (median) | | --- | ---: | ---: | ---: | | Light UI | 3 | 13,852 / 0.361 / 286.4 | 34,595 / 0.901 / 41.6 | | Light UI | 9 | 11,515 / 0.300 / 13937.6 | 12,687 / 0.330 / 406.9 | | Labelled drawing | 3 | 3,554 / 0.145 / 98.9 | 8,324 / 0.339 / 21.8 | | Labelled drawing | 9 | 1,644 / 0.067 / 7716.8 | 2,208 / 0.090 / 188.2 | | Painting | 3 | 150,434 / 6.121 / 135.4 | 140,851 / 5.731 / 44.1 | | Painting | 9 | 119,700 / 4.871 / 28956.9 | 121,901 / 4.960 / 1320.4 | | Photograph | 3 | 290,619 / 7.568 / 159.1 | 274,218 / 7.141 / 62.8 | | Photograph | 9 | 258,681 / 6.736 / 32734.2 | 254,579 / 6.630 / 2132.3 | | Terminal | 3 | 9,799 / 0.399 / 148.7 | 40,983 / 1.668 / 28.2 | | Terminal | 9 | 7,814 / 0.318 / 13108.0 | 11,729 / 0.477 / 246.3 |

Node 22.23.3 on Linux x64, AMD EPYC 9V74, CPU 0. One encoding thread per implementation. Rapier reports one call per configuration without warmup; libjxl reports the median of three calls after one warmup. Rapier uses automatic WASM; libjxl is native C++. The two implementations were measured separately on the same idle host. Timing includes each encode and output, excluding input transport/loading and decoding. Worker startup was 38.3 ms, measured separately. Every result preserves exact source RGBA; Rapier outputs also pass two independent decoders.

At effort 9, Rapier produces smaller files on these three document images and the painting; libjxl produces the smaller photograph. Native libjxl is faster in every measured row. Rapier's extra effort-9 search trades encoding time for smaller files and retains every earlier candidate. These desktop measurements do not predict browser or phone performance.

All four efforts, individual samples and reproduction.

Use the complete worker

Copy node_modules/rapier-jxl/dist/rapier-worker.js beside your page:

const worker = new Worker('./rapier-worker.js');
worker.onmessage = ({data}) => {
  if (data.ok) console.log(new Blob([data.bytes], {type: 'image/jxl'}));
  else console.error(data.error.message);
};
worker.postMessage({
  id: 1, operation: 'encode', width: 1, height: 1,
  data: new Uint8Array([255, 0, 0, 255]), options: {lossless: true}
});

For one offline HTML file, embed that complete worker as text and create it through a Blob URL. Copyable worker and module examples. Keep the MIT notice. No network download or decoder is needed for encoding.

The core and effort modules expose encode(rgba, width, height, options). Quality 100 preserves every sample, including RGB under transparent pixels; 1–99 is lossy. Alpha stays exact. Core/effort default to quality 100, and the separate effort entry defaults to effort 1. API and options.

Native inputs include 8/10/12/16-bit integers, binary16/32 floats, PQ/HLG and Rec. 2020. rapier-jxl/source reads supported PNG16/OpenEXR files. Displaying encoded output requires JPEG XL support.

Agent skills: Single-file apps · Photography.

API · Entry points and embedding · Agent reference · Comparisons · Browser example