Free diff checker — paste two versions and every change is marked, line by line and word by word.
No sign-up and no size limit worth worrying about; your text is compared right here and never sent anywhere.
Put the original on the left and the changed version on the right. The comparison runs as you type, so there is nothing to press. That is the whole of it — a difference checker online should not need a manual. The same operation covers a lot of ground: a diff checker tool for source, a data diff checker for CSV and JSON exports, and a plain difference text checker when what you are comparing is prose.
.diff file.Three states, and each answers a different question when you check difference text between two versions:
| Colour | Meaning | Where it appears |
|---|---|---|
| Red | Line is on the left but not the right | Original only |
| Green | Line is on the right but not the left | Changed only |
| Amber | Same line, edited | Both sides, with the changed words darker |
| Plain | Identical | Both sides |
That last one is the part most tools skip. Telling you a line changed is easy; showing you which three words changed inside a forty-word paragraph is the difference between a result you can read and one you have to squint at. Tick compare by character and it narrows further still, down to individual letters — useful for a changed number or a single typo.
Value and value as the same. Useful for config keys and SQL, misleading for anything case-sensitive.
Drop a file on either box and it loads straight into it. This is where a
document difference checker usually asks you to sign up; here the file is
read by the page and never leaves the machine. Anything plain text is fair game, which in
practice covers most of what people actually want to
diff check between two files: source in any language, JSON and XML
payloads, CSV exports, SQL dumps, server logs, .env and config files,
Markdown drafts, subtitle files.
Long files are handled by skipping the identical stretches at the beginning and end before the comparison starts, which is why a five thousand line file with one changed line comes back instantly. Two very large files that differ almost everywhere hit a deliberate ceiling and fall back to a coarser answer — and the page says so rather than pretending.
Most people reach a diff checker holding something they should not be handing to a stranger: an employment contract, a production config with credentials in it, source code that belongs to an employer. The usual tools send all of it to a server to do the work.
This one does not. The algorithm, the highlighting and the download all run in the page itself, so your text never crosses the network. It also means the tool keeps working with the connection off once the page has loaded. If you have ever hesitated before pasting something into a comparison box, that hesitation was correct, and this is the answer to it.
Worth saying plainly. It compares plain text. It will not compare Word documents, PDFs or spreadsheets: those store layout and styling alongside the words, so a diff of the raw file is unreadable and a diff of the extracted text quietly throws away the formatting that often was the change. It also does not merge, and it compares two things rather than three or a whole folder.
Those are deliberate limits, not missing features. A comparison you cannot trust is worse than no comparison, because a wrong answer looks exactly like a right one.