Quick summary: Smallpdf, Sejda, PDF24, and CommandPDF are all capable PDF suites, and for a casual merge or compress any of them will get the job done. The decisive difference between them is architecture: PDF24 and CommandPDF let you process files locally without uploading, while Smallpdf and Sejda process on their servers. That single fact explains the privacy posture, the file-size limits, the offline behavior, and a lot of the pricing. This is an honest, source-it-yourself comparison — including where the competitors are genuinely better.
The one comparison that matters most
Before features, limits, or price, there is a more basic question: does the tool need to receive your file?
| Tool | Where your file is processed | Your file ever leaves your device? |
|---|---|---|
| CommandPDF | Your browser (WebAssembly) | Never |
| PDF24 | Your browser (desktop/web) for most tools | No (for the local tools) |
| Smallpdf | Their servers | Yes |
| Sejda | Their servers | Yes |
This is not a marketing claim — it is verifiable. Open DevTools → Network in any of them and watch what happens when you run a job. CommandPDF and PDF24's local tools issue no upload request; Smallpdf and Sejda upload your file to process it. (For a full walkthrough, see Client-Side vs Server-Side PDF Processing.)
Everything below is a consequence of that architectural choice.
Feature and limit comparison
| CommandPDF | PDF24 | Smallpdf | Sejda | |
|---|---|---|---|---|
| Architecture | Client-side | Mixed (desktop + web, many local) | Server-side | Server-side |
| Approx. tool count | 90+ | 30+ | ~20 | ~30 |
| Edit PDF text | Free | Free | Free (limited) | Free (limited) |
| Merge / split / compress | Free | Free | Free (with limits) | Free (with limits) |
| OCR | Free | Free | Paid | Paid |
| Redaction (true removal) | Free | Limited | Paid | Paid |
| Digital signatures (X.509) | Free | Desktop only | Limited | Paid |
| Workflows (chain tools) | Free | No | No | No |
| Account required | Never | No (for local tools) | For most things | For most things |
| Watermark on output | Never | No | No | No |
| Works offline after load | Yes | Yes (desktop/local) | No | No |
| File-size limit | Your device's memory | Your device (local) | Free-tier limit | Free-tier limit (~200 MB / 50 MB) |
| Price model | Free | Free | Free + paid Pro | Free + paid Pro |
Note: exact free-tier limits and Pro prices change frequently — check each site for current numbers. The pricing model (free-with-quotas vs fully free) is the durable signal, and that follows architecture: server-side tools have real per-job compute and storage costs, which is why they gate usage behind quotas and paid tiers. Client-side tools spend your device's resources, not the operator's, so there's nothing to meter.
Where each tool genuinely wins
An honest comparison means naming where the competitors are better.
Smallpdf has the slickest onboarding of the four — the fastest path from "I have a PDF" to "I have a result" for a casual user who doesn't mind uploading. The design polish is real, and for non-sensitive files the friction is the lowest in the group.
Sejda has a deep, mature editing surface — particularly for direct in-document text editing and granular page operations — that reflects years of focused work. For heavy editing of a non-sensitive document, it's a strong choice.
PDF24 deserves special credit here: like CommandPDF, it offers genuinely free, locally-processing tools (its desktop app and many of its web tools don't require uploading), and it has been a privacy-conscious option in this space for a long time. If you prefer a desktop install and a German-data-protection posture, PDF24 is the obvious alternative to CommandPDF.
CommandPDF wins on three axes: the privacy guarantee is structural (no upload, ever, including for redaction and encryption of sensitive material); the toolset is the widest of the four (90+ tools, including redaction, digital signatures, PDF/A archival, and a workflow editor for chaining operations); and there are no quotas, accounts, or premium gates — every tool is free because there's no server cost to recover.
Privacy ranking
For a document you'd rather not have sitting on a third-party server — a contract, an ID, a medical or financial record — the ranking is objective:
- CommandPDF / PDF24 (local tools) — your file does not leave your device. Nothing to retain, breach, or compel.
- Smallpdf / Sejda — your file is uploaded and processed server-side, then deleted per policy. The protection is a promise about retention, not a structural guarantee.
This isn't about which company is more trustworthy; it's that the server-side model carries a category of risk the client-side model structurally cannot have. See When "Free" PDF Tools Cost You Everything for the full breakdown of those failure modes.
Which should you actually use?
| Use case | Better choice |
|---|---|
| Confidential document (contract, ID, medical, financial) | CommandPDF or PDF24 (local) |
| Everyday merge / compress / sign of non-sensitive files | Any — CommandPDF starts fastest (no upload) |
| Heavy direct text editing of a non-sensitive PDF | Sejda (mature editor) |
| Slickest casual onboarding, non-sensitive files | Smallpdf |
| You want a desktop app, EU-hosted | PDF24 |
| You want redaction, encryption, or digital signing of sensitive material | CommandPDF |
| Working offline | CommandPDF or PDF24 |
Many people keep two of these bookmarked and pick based on the sensitivity of the document. That's a perfectly reasonable approach — use whatever is convenient for low-stakes files, and reach for the architecture that matches the stakes when the document matters.
Frequently asked questions
Why are CommandPDF and PDF24 free without quotas while the others aren't? Because of where the work happens. Server-side processing costs the operator real money per job (compute, storage, bandwidth), so they meter it with quotas and paid tiers. Client-side processing runs on your device's resources, so there is nothing to meter and nothing to charge for.
Is "free" client-side processing worse quality? No. CommandPDF uses the same families of industry-standard PDF engines (MuPDF, pdf-lib, PDF.js) that server-side products use. Output correctness depends on the engine, not on where it runs.
If I don't care about privacy, why pick CommandPDF? Two reasons that hold even for non-sensitive files: there's no upload/download wait, so jobs start instantly; and there are no quotas, accounts, or daily limits to hit. The privacy guarantee is a bonus, not the only reason.
Aren't server-side tools faster for big jobs? They can be, for the heaviest operations (large OCR, big office conversions) on weak hardware, because a powerful server CPU beats an old laptop. For everything else, skipping the upload makes client-side as fast or faster in practice. We cover this honestly in Client-Side vs Server-Side PDF Processing.
Conclusion
These four tools overlap heavily on features and diverge sharply on architecture. If your files are casual, pick by convenience and editing polish. If your files are sensitive, pick by where they're processed — and that narrows the field to the client-side options. CommandPDF is built for the second case, and happens to be competitive on the first.
Try CommandPDF — 90+ tools, all in your browser, all free →
Related reading: