wkhtmltopdf Is Archived. Here's the Modern Replacement.

If you're using wkhtmltopdf in production, you've probably noticed: it was archived in 2023. No more updates. No bug fixes. And it's been struggling for years before that — broken CSS Grid support, no JavaScript execution, install nightmares on modern Linux.

This guide covers why developers are replacing wkhtmltopdf with a hosted PDF API, and how to switch your existing code to URLSnap in minutes.

Why wkhtmltopdf Breaks on Modern Stacks

wkhtmltopdf was built on an old Qt WebKit engine — essentially Internet Explorer-era browser technology. By modern web standards, it's severely limited:

wkhtmltopdf vs URLSnap: Side-by-Side

Migration: From wkhtmltopdf to URLSnap

Here's a typical wkhtmltopdf shell call or Node.js wrapper:

# Before: wkhtmltopdf shell command
wkhtmltopdf --page-size A4 https://example.com output.pdf

# Before: Node.js with wkhtmltopdf wrapper
const wkhtmltopdf = require('wkhtmltopdf');
wkhtmltopdf('https://example.com', { pageSize: 'A4' })
  .pipe(fs.createWriteStream('output.pdf'));

Here's the equivalent with URLSnap:

# After: URLSnap (curl)
curl "https://urlsnap.dev/api/pdf?url=https://example.com&format=A4&apiKey=YOUR_KEY" \
  -o output.pdf

// After: Node.js with URLSnap
const response = await fetch(
  `https://urlsnap.dev/api/pdf?url=https://example.com&format=A4&apiKey=YOUR_KEY`
);
const buffer = await response.arrayBuffer();
fs.writeFileSync('output.pdf', Buffer.from(buffer));

No binary to install. No library conflicts. No ARM compatibility issues. One HTTP request from any language, anywhere.

Python Migration

# Before: Python + pdfkit (wkhtmltopdf wrapper)
import pdfkit
pdfkit.from_url('https://example.com', 'output.pdf')

# After: URLSnap
import requests
response = requests.get(
    'https://urlsnap.dev/api/pdf',
    params={'url': 'https://example.com', 'format': 'A4', 'apiKey': 'YOUR_KEY'}
)
with open('output.pdf', 'wb') as f:
    f.write(response.content)

PHP Migration

// Before: PHP + wkhtmltopdf
$pdf = new Knp\Snappy\Pdf('/usr/local/bin/wkhtmltopdf');
$pdf->generate('https://example.com', 'output.pdf');

// After: URLSnap
$response = file_get_contents(
    'https://urlsnap.dev/api/pdf?url=https://example.com&format=A4&apiKey=YOUR_KEY'
);
file_put_contents('output.pdf', $response);

Works Everywhere wkhtmltopdf Doesn't

• Docker: No apt packages, no library pinning. URLSnap is a REST call — it works in any container, including Alpine with zero setup.
• AWS Lambda: wkhtmltopdf binaries are too large for Lambda layers and fail on Amazon Linux 2023. URLSnap works from any Lambda function with zero setup.
• Vercel / Netlify: No binaries allowed. URLSnap is just fetch().
• Apple Silicon (M1/M2/M3): wkhtmltopdf has no native ARM builds. URLSnap runs on your Mac dev machine, your Graviton CI, and your production server identically.
• Any language: wkhtmltopdf wrappers exist for Node.js, Python, PHP, and Ruby — but they all shell out to the binary. URLSnap is a pure REST API, callable from Go, Java, C#, Rust, or anything that speaks HTTP.

URLSnap PDF API Reference

GET https://urlsnap.dev/api/pdf

# Required
url=https://example.com     # URL to convert to PDF
apiKey=YOUR_API_KEY

# Optional
format=A4                   # A4 (default), Letter, Legal, A3, A5, Tabloid
landscape=true              # landscape orientation
delay=1000                  # wait N ms after page load (for JS-rendered content)

Pricing

URLSnap is cheaper than maintaining your own wkhtmltopdf install — especially when you factor in the ops time:

• Free: 20 requests/day — no credit card, just sign up and start generating PDFs
• Starter ($9/mo): 500 req/day — more than enough for most apps and SaaS products
• Pro ($29/mo): 5,000 req/day — production workloads, invoice generation at scale

The free tier alone is enough to evaluate the output quality and prove your migration works before committing a cent.

FAQ

Does URLSnap execute JavaScript like a real browser?

Yes. URLSnap uses headless Chromium — a full modern browser engine. JavaScript runs, CSS Grid and Flexbox render, web fonts load, and SPAs built with React or Vue render correctly. Use ?delay=2000 if your page has async data loading that needs time to complete before PDF generation.

Can I generate PDFs from HTML strings, not just URLs?

The current API accepts URLs. For HTML-to-PDF from a string, host your HTML template on your own server (even a local one tunneled via ngrok during development) and pass the URL to URLSnap. This keeps the API simple and stateless.

What about wkhtmltoimage for screenshots?

URLSnap covers that too — the /api/screenshot endpoint converts any URL to PNG or JPEG, with full JavaScript rendering. It's a complete replacement for both wkhtmltopdf and wkhtmltoimage.

What if my pages need authentication?

For publicly accessible pages, URLSnap works perfectly. For pages behind login, generate a one-time-use public URL or use a pre-signed token in your URL. Pages that can be accessed without credentials are the primary use case.

Is there a wkhtmltopdf wrapper library for URLSnap?

You don't need one — URLSnap is a plain REST API, far simpler than wkhtmltopdf's CLI interface. A single fetch() or requests.get() call is all you need in any language.