Validate

Compare two blocks of text and highlight every difference

Compare two versions of a text document and highlight additions, deletions, and changes.

🛡️ Your rows never leave your browser.

Why Onfyt: Word-level diff highlights the specific word within a changed line that differs, not just the whole line — useful for spotting a single changed number or name in a long sentence.

Interactive Sample Data

id	text_a	text_b
1	Meeting on Monday.	Meeting on Tuesday.
2	Attendees: Grace, Peter, Alice.	Attendees: Grace, Peter, Bob.

How this tool works

Comparing two versions of a contract, configuration file, email template, or data export by eye is error-prone — a single changed word in a 50-line document is easy to miss. A diff tool shows exactly which lines were added, removed, or changed, and highlights the specific words or characters that differ. This tool produces a unified diff output — the same format used by Git, SVN, and code review systems — decorated with colour for additions (green), deletions (red), and unchanged context lines (grey). It accepts two text blocks: - **Text A** (the original): paste the original version in the first part of your instruction or label it "original:". - **Text B** (the revised): paste the revised version and label it "revised:". Diff modes available through your instruction: - **Line diff** (default): reports which whole lines were added, removed, or changed. - **Word diff**: within changed lines, highlights the specific words that differ. - **Ignore whitespace**: leading/trailing space differences are not counted as changes. - **Ignore case**: 'Hello' and 'hello' are treated as identical. All comparison runs inside your browser. Neither text block reaches any server.

Things that usually go wrong

If both texts are pasted as one block without labels, the engine may not split them correctly — always separate with 'original:' and 'revised:' labels or a clear dividing line.
Very similar texts where every line changed slightly will produce a long diff report — use word-level mode to see only the specific characters that differ.
Diff of code blocks with inconsistent indentation will flag every line as changed — normalise indentation first.

Edge cases we tested

🧪 Identical texts

Returns '0 differences found — texts are identical.' rather than an empty diff block.

🧪 Only whitespace differences (trailing space added on each line)

Flagged as changes by default; ignored when 'ignore whitespace' is specified.

🧪 One text is empty, the other is not

Every line of the non-empty text is shown as an addition.

🧪 Texts with different line endings (CRLF vs LF)

Normalised before comparison — line-ending differences are not reported as content changes.

🧪 Very long lines (1000+ characters per line)

Word-level diff still works; the output wraps at the display boundary.

Frequently Asked Questions

Can I compare CSV files rather than plain text?

Yes — paste both CSV files and the diff shows which rows changed. For row-level CSV comparison, use 'Compare Two JSON Arrays' or 'Find Duplicate Rows'.

Is the comparison case-sensitive by default?

Yes. Add 'ignore case' to your instruction to treat uppercase and lowercase as identical.

Can I get the output as a downloadable file?

Yes — the diff output can be downloaded as a .txt or .html file from the export menu.

What does 'unified diff format' mean?

It is the standard format used by Git pull requests: lines starting with + are additions, lines starting with - are deletions, and lines without a prefix are unchanged context.