Skip to content
1001 Tools
sr
Text tools

Text Difference Checker

Paste the original on the left, the new version on the right, and the differences appear below: green for added lines, red for removed, everything else confirmed unchanged. It is the same longest-common-subsequence algorithm behind the classic diff command — running privately in your browser.

Paste both versions to see the differences.

How it works

The comparison works line by line. The algorithm finds the longest sequence of lines the two texts share, then presents everything outside that as additions (present only in the new text) or removals (present only in the original). A modified line therefore shows as its old version removed plus its new version added.

Whitespace counts: a trailing space makes lines different, which is honest but sometimes surprising. If you only care about wording, run both texts through Remove Extra Spaces first and diff the cleaned versions.

Each side accepts up to 3 000 lines — the exact LCS algorithm needs memory proportional to the product of both lengths, and the limit keeps the page responsive on modest devices. That covers contracts, articles and most config files comfortably.

Practical examples

What changed in the contract?

The counterparty returned “the same document with small edits”. Paste both versions and know in seconds that two payment-term lines changed and a penalty clause was added.

Comparing config file versions

The staging server works and production doesn’t. Diff the two .env or nginx files — the one changed line glows red/green instead of hiding among 200 identical ones.

Reviewing a colleague’s edit of your text

An editor returned your article “with light touches”. See exactly which sentences were rewritten before accepting — and which paragraph quietly disappeared.

Verifying a copy-paste survived intact

After moving a long list between systems, diff source against destination. Identical texts are confirmed explicitly — “texts are identical” — not just by absence of red.

Frequently asked questions

Why does a changed line show as one removed plus one added?

The comparison unit is the whole line — there is no in-line character diff (yet). A line that changed is, from the algorithm’s view, an old line that vanished and a new one that appeared. Reading a red/green pair together shows you the edit.

The texts look identical but the diff shows differences — why?

Almost always invisible characters: trailing spaces, non-breaking spaces from Word, or Windows vs Unix line endings. Clean both sides with Remove Extra Spaces and diff again — if differences remain, they are real.

Does the order of the two texts matter?

Only for labeling: swapping sides turns every “added” into “removed” and vice versa. Convention here matches diff tools everywhere — original left, new version right.

Can it compare Word documents or PDFs?

Paste their text content, yes — formatting, images and tables are outside its scope. Copy each version into a side; the text layer is what gets compared.

Why is there a 3 000-line limit?

The exact algorithm builds a table of size lines×lines — at 3 000×3 000 that is 9 million cells, still instant; at 50 000 lines it would freeze the tab. Within the limit you get a guaranteed-optimal diff rather than a heuristic one.

Is moved text detected as “moved”?

No — a paragraph that moved shows as removed in one place and added in another. Detecting moves is a heuristic layer some tools add; this one keeps the honest, deterministic core.

Are my documents uploaded for comparison?

No. Both texts stay in your browser’s memory and the algorithm runs locally — contracts and unpublished drafts included. Nothing is sent, cached or logged.

Can I get a unified diff file for git or email?

Not yet — the output is visual. Producing patch-format output is a sensible future addition; for now, screenshot or copy the relevant lines.

Related tools