A wkhtmltopdf alternative for modern HTML

wkhtmltopdf served a generation of apps well. But its GitHub repository was archived in January 2023, and it renders with a QtWebKit engine from the mid-2010s. Flexbox is unreliable, CSS grid and CSS variables don't work, modern JavaScript fails silently, and nobody ships security fixes anymore.

Symptoms you've outgrown it

• A layout that looks right in the browser comes out stacked or overlapping in the PDF (flexbox/grid).
• Charts or templates built with modern JS render blank, because the old engine doesn't understand ES2015+ syntax.
• You pin an old Docker base image just to keep the patched-Qt build working.
• Security reviews flag it: an unmaintained engine that fetches HTML from user input is an SSRF and local-file-read risk.

The replacements, honestly compared

Option	Good for	Trade-off
Puppeteer / Playwright page.pdf()	Full control, same engine as Chrome	You run and patch headless Chromium (~300MB, memory spikes, fonts)
Gotenberg (Docker)	Self-hosted HTTP API around Chromium and LibreOffice	One more service to deploy, scale and update
WeasyPrint (Python)	Excellent CSS paged media: running headers, page counters	Its own engine: no JavaScript, some layout differences from browsers
A hosted API (Shotmint, PDFShift, …)	No browser to operate, one HTTP call	Your HTML goes over the network to a third party; costs per document

If you need JavaScript-rendered content and browser-identical layout, choose a Chromium option. If you need advanced print features (running headers, footnotes, counter(page)), WeasyPrint is hard to beat.

Before and after

# Before: wkhtmltopdf
wkhtmltopdf --page-size A4 -O Landscape --print-media-type input.html out.pdf

# After: one HTTP call (Chromium)
curl -X POST https://shotmint.creasework.com/v1/pdf \
  -H "X-API-Key: $SHOTMINT_KEY" -H "Content-Type: application/json" \
  --data-binary @<(jq -Rs '{html: ., format: "A4", landscape: true}' input.html) \
  -o out.pdf

Python (replacing pdfkit)

# Before
import pdfkit
pdfkit.from_string(html, "out.pdf", options={"page-size": "Letter", "margin-top": "15mm"})

# After
import os, requests
r = requests.post("https://shotmint.creasework.com/v1/pdf",
    headers={"X-API-Key": os.environ["SHOTMINT_KEY"]},
    json={"html": html, "format": "Letter", "margin": "15mm"}, timeout=60)
r.raise_for_status()
open("out.pdf", "wb").write(r.content)

Node.js (replacing wkhtmltopdf / html-pdf wrappers)

const r = await fetch("https://shotmint.creasework.com/v1/pdf", {
  method: "POST",
  headers: { "X-API-Key": process.env.SHOTMINT_KEY, "Content-Type": "application/json" },
  body: JSON.stringify({ html, format: "A4", margin: "10mm" }),
});
if (!r.ok) throw new Error(await r.text());
require("fs").writeFileSync("out.pdf", Buffer.from(await r.arrayBuffer()));

Option mapping

wkhtmltopdf	Shotmint /v1/pdf
--page-size A4	format=A4 (A3, A4, A5, Letter, Legal, Tabloid)
-O Landscape	landscape=true
-T -B -L -R 10mm	margin=10mm (all sides). For different sides, pass margin=0mm and set @page { margin: … } in CSS
--no-background	print_background=false
--zoom 0.8	scale=0.8
default (screen CSS)	screen_media=true
--print-media-type	default (print CSS)
--javascript-delay 1000	delay=1000, or wait_until=networkidle
--custom-header User-Agent …	user_agent=…
--header-html, --footer-html, --toc	Not supported. Put the header in your HTML; a <thead> repeats on every page

Migration gotchas

• Default media flips: wkhtmltopdf uses screen CSS unless you pass --print-media-type. Chromium uses print CSS. If the output suddenly hides elements, check your @media print rules or pass screen_media=true.
• Scaling differs: wkhtmltopdf's "smart shrinking" made pages look smaller. In Chromium, 1 CSS px = 1/96 inch, so set font sizes in pt or tune scale.
• Local files: file:// images and stylesheets can't be reached from a hosted renderer. Use absolute https:// URLs, or inline them as data URIs and <style>.
• Page breaks: the old page-break-* properties still work; the modern ones are break-before: page and break-inside: avoid. See the HTML invoice guide for print CSS tips.

Want to see the output first? Paste HTML into the free HTML to PDF converter (no key needed).

What it costs

One PDF = one credit, whatever the page count. €5 buys 2,000 PDFs, credits never expire and there's no subscription. Failed renders are refunded automatically. Compare that with running a Chromium container around the clock for a few hundred documents a month.