pdf-lib is a pure JavaScript library for creating and modifying PDFs at the document-object level. It can open an existing file, fill its form fields, copy pages between documents, and save it back with the rest of the structure intact. Polotno SDK is a design editor you embed in your app: it parses a PDF into a design schema, lets a person edit it on a canvas, and writes a fresh document. The question that decides it: what must survive the round trip?
Pick the one that matches what must survive
Pick pdf-lib if…
- Form fields must keep working. Reading, filling, and flattening AcroForms is a first-class feature in pdf-lib and unsupported in Polotno.
- Bookmarks, links, and metadata must survive. pdf-lib rewrites the file from its objects and keeps that structure, though by default it updates the producer and modified date. A schema import rebuilds the page and drops the structure.
- You are splitting documents or reusing pages. pdf-lib copies pages between files without reinterpreting their content. Copy into a new document and bookmarks and form fields stay behind.
- You want an open-source library with no license fee. It is MIT, and it runs in Node, the browser, Deno, and React Native.
- Fidelity to the original beats editability. Nothing is re-rendered, so nothing shifts.
Pick Polotno SDK if…
- A person needs to edit the document visually. pdf-lib has no editing surface; the canvas is the whole point here.
- The text needs to be editable. pdf-lib can draw new text onto a page but does not reflow or restyle the text already there. A schema import turns it into text elements you can change.
- You are generating from templates and data. One design, many rows, rendered through the Cloud Render API or on your own servers.
- You need formats beyond PDF. PNG, JPEG, SVG, HTML, PPTX, GIF, and MP4 from the same schema, in the browser or on your own servers.
- You need print output. CMYK, spot colors, bleed, crop marks, and PDF/X-1a or PDF/X-4 from PDF export.
Preserving a file vs reinterpreting it
1. What "edit" means
pdf-lib edits the document. The original content streams stay as they are and you modify objects around them. Polotno reinterprets the document as a design, which is why the text becomes editable and also why the structure around it does not come along.
2. The round trip
Run a PDF through pdf-lib and out again and you keep its structure and page content in a rewritten file. In our test on 1.17.1, outlines, form fields, tags, and link annotations all survived and the page content stayed byte-identical, while the file as a whole changed. Run one through Polotno and you get a new file rendered from a schema. That is a feature when the goal is to change the content and a liability when the goal is to preserve the artifact.
3. Merging
Both can merge, and they mean different things by it. pdf-lib copies pages structurally. In our test, a copy into a new document kept link annotations and lost bookmarks, form fields, and tags. Polotno's merge concatenates parsed pages into one design, preserving each page's own size but not its form fields, bookmarks, or links. If a merged file must keep bookmarks and forms, neither one does that for you.
Feature comparison
| Capability | Polotno SDK | pdf-lib |
|---|---|---|
| What it is | Design editor you embed, plus rendering | JavaScript library with no UI |
| Existing form fields | None: field text imports as static text | Reads, fills, and flattens AcroForms (no XFA) |
| Bookmarks and links on save | Dropped | Kept |
| Original page content | Rebuilt as a design | Left unchanged |
| Editing existing text | Text elements on a canvas | Not supported |
| Export formats | PNG, JPEG, PDF, PPTX, GIF, MP4, plus SVG and HTML outside Cloud Render | |
| License | Commercial, fixed license fee by custom quote, Cloud Render billed per use | MIT, no license fee |
| Latest release | Ongoing, see the changelog | 1.17.1, November 2021 |
When pdf-lib is the right call
Anything where the input file's integrity matters. Filling an AcroForm, stamping a header onto documents you did not author, splitting a batch of statements. In all of those, reinterpreting the page would be the wrong move, and pdf-lib leaves the page alone, which is why it fits.
It is also the right call when you need a small, free library and the task is mechanical.
Check two things before you commit. pdf-lib's docs say it does not support XFA forms or encrypted PDFs. And its latest release is 1.17.1 from November 2021, with no commit on its main branch since that month. The repository is not archived, and @cantoo/pdf-lib is an MIT fork with 2026 releases, so compare the two before you start something new.
If a person needs to edit the page visually, evaluate Polotno with the 30-day free trial, no credit card needed.
A naming ambiguity worth knowing
Searches for "pdflib" hit two unrelated products: pdf-lib, the MIT JavaScript library this page compares against, and PDFlib, a long-standing commercial C and C++ library with bindings for several languages. If you are on a JVM or .NET stack and someone recommends "PDFlib", check which one they mean.
FAQ
What can shift when Polotno imports a PDF?
Unusual embedded fonts and complex transparency can shift, and complex scripts such as Arabic import as vector artwork instead of editable text.
Can I use both?
Yes, and it is a sensible split. Use Polotno for the authoring and design surface, and pdf-lib for structural work on files you did not create.
Does Polotno compress PDFs?
No. Re-rendering through a schema is not a file-size optimization pass and should not be treated as one.
