WEBRuns in the browser — no install

Online plagiarism checker.

Paste, upload, or fetch a URL. 2,500 words free against 150M+ web and academic sources. Browser-native — no software install, no extension. Your history is there on any device you sign in from.

0 / 1,500 words
Any browser, any deviceCross-device history syncNo install · no extension
INPUT METHODS

Four ways to get text into the checker.

All of it stays in the browser session until you hit Check. Nothing transmitted on paste, nothing cached without consent.

01 · PASTE

Paste right into the box

Cmd+V (or right-click) anywhere in the textarea. Word-count updates as you type. The Free tier lets you check 2,500 words in one paste; Premium raises it to 50,000.

txtrich textany encoding
02 · UPLOAD

Drop a file from your device

Drag .docx / .pdf / .odt / .rtf / .txt straight onto the textarea. We extract the body text in the browser before any network call — your file metadata never leaves the page.

.docx.pdf.odt.rtf
03 · URL FETCH

Submit a public URL

Paste any public web URL — blog post, journal page, archived article. The browser fetches it server-side (so your IP isn't exposed to the source) and the readable body gets checked.

httphttpsarchive.org
04 · DRIVE / DROPBOX

Pull from your cloud drive

Connect Google Drive, Dropbox, or OneDrive once. Pick the document from a familiar file-picker, no re-upload. OAuth tokens stay on-device — we never see your full drive.

DriveDropboxOneDrive
BROWSER COMPATIBILITY

Tested every release against the four browsers your users actually run.

Modern engine baseline. Mobile parity isn't an afterthought — the textarea, drop zone, and report view all work on a phone.

BROWSERPASTEUPLOADURL FETCHSYNCPDF EXPORT
Chrome120+
Safari17+
Firefox121+
Edge120+
MobileiOS 16 · Android 12share-sheet
TabletiPadOS · Android
Tested in every release. Last regression suite: 2 days ago.
REPORT VIEWS

One check. Six ways to read the result.

Pick the view that matches what you're doing — proofreading the highlighted text, scanning a source list, exporting for a hand-in. All in the browser, all responsive.

Highlighted inline

Match colours overlay the original text — hover any segment to pop a source citation, click to jump to its position in the source list.

Source list panel

Sortable by similarity, by overlap length, by source domain. Filters for web / academic / Noplag DB / your folders. Click a source to see all intervals matching it.

Side-by-side compare

Your text on the left, the matched source excerpt on the right, scrolling locked together. Best for verifying paraphrase claims by eye.

Mobile compact

Phone-sized layout: similarity number up top, expandable per-source cards below. Pinch-zoom on the source text. Same engine, just laid out for the screen.

PDF export

Two-click export. Cover page with similarity score + engine commit, then the full report. Print-ready, hand-in-ready, defensible in a hearing.

Annotation overlay

Write your own notes against any matched interval. Notes save to your synced history — so when you reopen on a different device, your annotations are there.

HISTORY ACROSS DEVICES

Your checks, on every device you sign in from.

Your check history lives in your account rather than on one machine. Sign in on a new phone and your recent checks are already there. Documents are encrypted in transit and at rest, and are never used to train models.

MacBook Prothis device
  • Essay 4 — Modernist Novel23%4m ago
  • Cover letter — Pearson8%yesterday
  • Article draft — Substack31%2 days ago
  • Thesis chapter 314%5 days ago
iPhone 15signed in 6h ago
  • Essay 4 — Modernist Novel23%4m ago
  • Cover letter — Pearson8%yesterday
  • Article draft — Substack31%2 days ago
  • Thesis chapter 314%5 days ago
iPad Airsigned in last week
  • Essay 4 — Modernist Novel23%4m ago
  • Cover letter — Pearson8%yesterday
  • Article draft — Substack31%2 days ago
  • Thesis chapter 314%5 days ago
Encrypted in transit and at rest · never used to train models · EU data residency available on paid plans
  • 4browsers tested per release
  • .docx .pdf.odt .rtf .txt
  • <5ssync to new device
  • 0installs · 0 extensions
  • TLS 1.3encrypted in transit and at rest
PRIVACY ON A SHARED COMPUTER

Three things the browser does that a desktop installer can't.

01 · SIGNED-OUT MODE

Run a check without an account

The Free tier doesn't require signup. On a borrowed laptop or library terminal, paste 2,500 words, see the report, close the tab. No session left behind, no history written, nothing to remember to log out of.

Nothing persisted server-side beyond rate-limit counters (hashed IP, 24h).
02 · LIBRARY / KIOSK

The browser doesn't install anything

A school library or shared workstation often blocks .exe installs and extensions by policy. Noplag runs entirely in the browser tab — no Chrome extension, no native helper app, no admin permission needed. Closes when the tab closes.

Tested on locked-down Chromebooks, RDP thin clients, Citrix sessions.
03 · INCOGNITO

Same engine, nothing left behind

Open it in Private / Incognito / InPrivate. The check runs against the same engine and corpus. Cross-device sync is disabled (no account = nothing to sync to), but the report is full-featured and exportable to PDF before you close the window.

Works in Safari Private Browsing, Firefox Private Window, Chrome Incognito.
NO INSTALL · NO EXTENSION

Twelve years of plagiarism checking — and we never built a desktop client.

Every other major plagiarism vendor ships some installer, extension, or LMS plugin you have to convince IT to approve. We never did, because the right shape for this product is a URL — that's what made the original noplag.com free version useful on a school computer in 2014, and it's what makes the rebuild useful on a phone in 2026. The full check happens server-side, the UI runs client-side; nothing in between you and the result needs admin permission, Gatekeeper approval, or SmartScreen consent. The closest thing we'll ever ship to an installer is docker compose up on your own server — and that's only because Apache 2.0 means you can if you want to.

Read why no installer
FAQ

The questions people ask before they paste.

Do I really not need to install anything?

Right. There's no Chrome extension, no desktop client, no LMS plugin requirement. The whole product is at noplag.com — paste, check, read the report, close the tab. The only “install” we ship is docker compose up for self-hosting the engine, which is optional.

What browsers and devices are supported?

Chrome 120+, Safari 17+, Firefox 121+, Edge 120+. Mobile: iOS 16+ and Android 12+. Tablet: iPadOS and Android tablet builds. The textarea, file-drop, and report view are all responsive — same engine, mobile-aware layout. Regression-tested on every release.

What file formats can I upload?

.docx, .pdf, .odt, .rtf, and plain text. Text extraction runs in the browser before any network call, so your file metadata (author, revision history, comments) never leaves the page. Up to 2,500 words on Free, 50,000+ on Premium.

What if my text is in another language?

Detection is calibrated for English, Spanish, Portuguese, Polish, and Ukrainian at launch, with more languages rolling out. The detection runs per-language with calibration validated against public benchmarks — not a one-size-fits-English model.

What's the privacy story on a shared computer?

Three layers: signed-out mode (the free check needs no account, so no session is left behind); private-window support (the full check and report work in Incognito or Private Browsing, and nothing survives closing the window); and analytics that stay off unless you accept them. Close the tab, you're done.

How fast does the check actually run?

Median 42 seconds for a 1,500-word document against the full corpus (web + academic + Wikipedia + Common Crawl). p95 is ~75s. Long documents (5,000+ words) sometimes need a webhook callback — the UI handles this without making you wait on a frozen tab.

Does the browser-only design limit accuracy?

No — the heavy lifting (retrieval cascade, alignment, dedupe) happens server-side. The browser is just the UI. Self-host the engine on your own server if you want the entire flow to stay inside your network; same algorithms either way.

What's the URL-fetch feature for?

Paste a public web URL — blog post, journal page, archived article — and the engine fetches it server-side and runs the check on the readable body. Useful for cross-checking a quoted source, or for SEO writers verifying a published competitor article. Your IP isn't exposed to the source.

Can I export the report?

PDF export in two clicks: cover page with similarity score + engine commit, then the full report. On mobile, share-sheet integration lets you AirDrop it or send to Files / Drive / Dropbox without leaving the browser.

How does my check history work across devices?

Your history is stored in your account, so signing in on another device shows the same checks. It is encrypted in transit and at rest, access is scoped to your workspace and logged, and your documents are never used to train models. We do not offer client-side encryption today — if you need us to be technically unable to read your documents, self-host the Apache 2.0 engine and nothing leaves your infrastructure.

Open the URL. Run a check. That's the whole setup.

Free tier covers 2,500 words against 150M+ web and academic sources. No card, no signup, no “download our app” prompt. Works on the phone you're holding right now.

Online Plagiarism Checker — Free, Browser-Based | Noplag