← All guides

How to convert PDF to ZPL

You have a PDF (usually a carrier shipping label from UPS, FedEx, USPS, Endicia or EasyPost) and a Zebra thermal printer that only speaks ZPL. Here are three ways to convert it, fastest first:

1. Free online converter - drop the PDF, get ZPL per page with a pixel-true preview. No signup; anonymous output carries a small stripyhorse.io logo in the corner, and a free account removes it.
2. Open-source PHP library - benfaerber/pdf-to-zpl, 26,000+ installs, runs inside your own app. Needs Imagick with the PDF coder enabled; if it refuses the file with a security policy error, that is your distro's policy.xml, not your code.
3. REST API - one curl (or a PHP/Python/C#/Java SDK call) from any stack, shown below.

An 8-page PDF converted by this engine and printed on a real Zebra; each page one label.

What a PDF→ZPL conversion actually does

ZPL has no PDF support. The conversion rasterizes each PDF page at the printer's exact density, converts the pixels to a 1-bit bitmap, and wraps it in a ^GFA (Graphic Field) command with Zebra's ACS run-length compression:

^XA
^GFA,121836,121836,102,,:::::N03FE,M01JF8...
^XZ

That blob is the label. Send it to the printer over TCP port 9100 and it prints. No drivers, no middleware:

The mistake that ruins most conversions: density

Zebra printers come in four densities, and ZPL coordinates are physical dots, so a label converted for the wrong density prints at the wrong size:

If your 4×6 label prints as a small rectangle in the corner, it was converted at 8 dpmm and printed on a 12 dpmm printer (or vice versa, cropped). Convert at the density of the target printer; don't rescale afterward, which blurs barcode edges.

Converting via the API

curl https://api.stripyhorse.io/v1/convert \
  -H "X-Api-Key: sh_live_YOUR_KEY" \
  -F file=@shipping-label.pdf -F preset=4x6 -F dpmm=8

Each page comes back as its own ZPL string. There is also a bulk endpoint that accepts up to 20 files and streams results per page as they finish, useful for end-of-day batch processing.

Will the barcodes still scan?

Yes, if two things are true:

1. Rasterize at final size in one pass. Rendering a PDF large and then shrinking it resamples the barcode's bars into gray edges, and the 1-bit threshold then rounds bar widths inconsistently. Scanners tolerate some of this; at small module widths they stop. A correct converter renders the PDF directly at the target pixel size.
2. The source PDF's barcode has enough resolution for your density. A barcode that's crisp at 300 dpi can alias at 152 dpi. When in doubt, print one and scan it, or use a converter that re-renders detected barcodes as native ZPL fields (^BC) at printhead precision instead of rasterizing them.

Our engine is cross-validated against thousands of physically printed and scanned carrier labels: the same pipeline as the open-source library, rewritten in Go and roughly 100× faster (~45 ms per page).

FAQ

Can I convert PDF to ZPL for free?

Yes. The online converter is free without signup, though anonymous output carries a small stripyhorse.io logo in the corner that a free account removes. The PHP library is MIT-licensed and unwatermarked.

Does the output work on any Zebra printer?

^GFA graphics print on every ZPL-II printer (ZD, ZT, GX, GK, 105SL…), and on most Zebra-compatible printers from Honeywell/Datamax with ZSim enabled. Match the dpmm.

How do I convert multi-page PDFs?

Each PDF page becomes one label. The API returns them in order; the bulk endpoint streams {"file":…,"page":…,"zpl":…} objects as pages finish.

PNG or JPG instead of PDF?

Same pipeline: the image converter accepts PNG, GIF and JPEG.

Published 2026-08-24 · Ben Faerber, software engineer, five years building warehouse fulfillment and label-printing systems; maintainer of pdf-to-zpl (27,000+ installs powering production label generation).