pdf-oxide-wasm

PDF Oxide for WASM — The Fastest PDF Toolkit for Browsers, Deno, Bun & Edge

The fastest WebAssembly PDF library for text extraction, image extraction, and markdown conversion. Powered by a pure-Rust core compiled to WebAssembly. Runs in Node.js, browsers, Deno, Bun, and serverless edge runtimes — no native binaries, no node-gyp, no postinstall. 0.8ms mean per document, 5× faster than PyMuPDF, 15× faster than pypdf. 100% pass rate on 3,830 real-world PDFs. MIT / Apache-2.0 licensed.

Part of the PDF Oxide toolkit. Same Rust core, same speed, same 100% pass rate as the Rust, Python, Go, JavaScript / TypeScript (Node.js native), and C# / .NET bindings.

Need a faster Node.js binding with native code? Use pdf-oxide instead — same API, native N-API addon.

Quick Start

npm install pdf-oxide-wasm

const { WasmPdfDocument } = require("pdf-oxide-wasm");
const fs = require("fs");

const bytes = new Uint8Array(fs.readFileSync("paper.pdf"));
const doc = new WasmPdfDocument(bytes);

console.log(doc.extractText(0));
console.log(doc.toMarkdown(0));

doc.free();

Why pdf-oxide-wasm?

Feature	pdf-oxide-wasm	pdf-parse	pdf-lib	pdfjs-dist
Text extraction	Yes	Yes	No	Yes
Markdown / HTML output	Yes	No	No	No
PDF creation	Yes	No	Yes	No
Form field read/write	Yes	No	Partial	No
Full-text search (regex)	Yes	No	No	No
Image extraction	Yes	No	No	No
Merge, encrypt, edit	Yes	No	Yes	No
Serverless / edge runtimes	Yes	No	No	No
Zero native dependencies	Yes	Yes	Yes	No
WebAssembly-based	Yes	No	No	No
TypeScript types included	Yes	No	Yes	Yes
License	MIT / Apache-2.0	MIT	MIT	Apache-2.0

• Fast — 0.8ms mean per document, 5× faster than PyMuPDF, 15× faster than pypdf
• Reliable — 100% pass rate on 3,830 test PDFs, zero panics, zero timeouts
• Universal — Runs in Node.js, browsers, Deno, Bun, and Cloudflare Workers without modification
• Zero install friction — No native binaries, no node-gyp, no postinstall scripts
• Pure Rust core — Memory-safe, panic-free, compiled straight to WebAssembly
• Full TypeScript support — Type definitions ship in the package

Performance

Benchmarked on 3,830 PDFs from three independent public test suites (veraPDF, Mozilla pdf.js, DARPA SafeDocs). Text extraction libraries only. Single-thread, 60s timeout, no warm-up.

Library	Mean	p99	Pass Rate	License
PDF Oxide	0.8ms	9ms	100%	MIT / Apache-2.0
PyMuPDF	4.6ms	28ms	99.3%	AGPL-3.0
pypdfium2	4.1ms	42ms	99.2%	Apache-2.0
pdftext	7.3ms	82ms	99.0%	GPL-3.0
pdfminer	16.8ms	124ms	98.8%	MIT
pypdf	12.1ms	97ms	98.4%	BSD-3

99.5% text parity vs PyMuPDF and pypdfium2 across the full corpus. The WASM compilation preserves near-native performance — no garbage collection overhead, no child process spawning, no temp files.

Installation

npm install pdf-oxide-wasm

Works without modification in:

• Node.js 18+ (CommonJS and ESM)
• Browsers — Chrome, Firefox, Safari, Edge
• Cloudflare Workers — runs in V8 isolates with WASM support
• Deno — native WASM support
• Bun — native WASM support

No native binaries, no system dependencies, no build step.

API Tour

Open and extract text

const { WasmPdfDocument } = require("pdf-oxide-wasm");
const fs = require("fs");

const bytes = new Uint8Array(fs.readFileSync("document.pdf"));
const doc = new WasmPdfDocument(bytes);

console.log(`Pages: ${doc.pageCount()}`);
console.log(doc.extractText(0));        // plain text
console.log(doc.toMarkdown(0));         // markdown
console.log(doc.toHtml(0));             // HTML

doc.free();

ESM / TypeScript:

import { WasmPdfDocument } from "pdf-oxide-wasm";
import { readFile } from "fs/promises";

const bytes = new Uint8Array(await readFile("document.pdf"));
const doc = new WasmPdfDocument(bytes);

const text = doc.extractAllText();
const markdown = doc.toMarkdownAll();

doc.free();

Search

const results = doc.search("quarterly revenue", true); // case-insensitive
// Returns: [{ page, text, bbox, start_index, end_index, span_boxes }]

Form fields

const fields = doc.getFormFields();
// [{ name, field_type, value, tooltip, bounds, is_readonly, is_required }]

doc.setFormFieldValue("name", "Jane Doe");
doc.setFormFieldValue("agree_terms", true);

const filledPdf = doc.saveToBytes();

Create a PDF from Markdown

import { WasmPdf } from "pdf-oxide-wasm";

const pdf = WasmPdf.fromMarkdown("# Invoice\n\nTotal: $42.00", "Invoice", "Acme Corp");
const bytes = pdf.toBytes();

Encrypt a PDF (AES-256)

const encrypted = doc.saveEncryptedToBytes(
  "user-password",
  "owner-password",
  true,  // allow print
  false, // deny copy
);

Render and extract images

const images = doc.extractImages(0);
const pngBytes = doc.extractImageBytes(0);

Edit metadata, pages, and content

doc.setTitle("Quarterly Report");
doc.setAuthor("Example Author");
doc.setPageRotation(0, 90);
doc.cropMargins(36, 36, 36, 36);
doc.eraseRegion(0, 50, 50, 200, 100);
doc.flattenAllAnnotations();

const editedBytes = doc.saveToBytes();

Other languages

PDF Oxide ships the same Rust core through six bindings:

• Rust — cargo add pdf_oxide — see docs.rs/pdf_oxide
• Python — pip install pdf_oxide — see python/README.md
• Go — go get github.com/yfedoseev/pdf_oxide/go — see go/README.md
• JavaScript / TypeScript (Node.js native) — npm install pdf-oxide — see js/README.md
• C# / .NET — dotnet add package PdfOxide — see csharp/README.md

A bug fix in the Rust core lands in every binding on the next release.

Documentation

• Full Documentation — Complete documentation site
• WASM Getting Started — Step-by-step WASM guide
• Main Repository — Rust core, CLI, MCP server, all bindings
• Performance Benchmarks — Full benchmark methodology and results
• GitHub Issues — Bug reports and feature requests

Use Cases

• Browser PDF tooling — Extract, search, and convert PDFs entirely client-side, no server upload
• Edge / serverless workers — Process PDFs in Cloudflare Workers, Vercel Edge, Deno Deploy
• RAG / LLM pipelines — Convert PDFs to clean Markdown for retrieval-augmented generation
• PDF generation — Create invoices, reports, certificates programmatically without a backend
• Universal Node.js packages — Same code runs in Node.js, the browser, and edge runtimes

Why I built this

I needed PyMuPDF's speed without its AGPL license, and I needed it in more than one language. Nothing existed that ticked all three boxes — fast, MIT, multi-language — so I wrote it. The Rust core is what does the real work; the bindings for Python, Go, JS/TS, C#, and WASM are thin shells around the same code, so a bug fix in one lands in all of them. It now passes 100% of the veraPDF + Mozilla pdf.js + DARPA SafeDocs test corpora (3,830 PDFs) on every platform I've tested.

If it's useful to you, a star on GitHub genuinely helps. If something's broken or missing, open an issue — I read all of them.

— Yury

License

Dual-licensed under MIT or Apache-2.0 at your option. Unlike AGPL-licensed alternatives, pdf_oxide can be used freely in any project — commercial or open-source — with no copyleft restrictions.

Citation

@software{pdf_oxide,
  title = {PDF Oxide: Fast PDF Toolkit for Rust, Python, Go, JavaScript, and C#},
  author = {Yury Fedoseev},
  year = {2025},
  url = {https://github.com/yfedoseev/pdf_oxide}
}

---

WASM + Rust core | MIT / Apache-2.0 | 100% pass rate on 3,830 PDFs | 0.8ms mean | 5× faster than the industry leaders