Compare two blocks of text side-by-side. Highlights added, removed, and unchanged lines with a visual diff view.
Diffing is a matching problem before it is a display problem. The Text Diff Checker builds a longest-common-subsequence table across the two versions, then walks it backwards to decide which lines are shared, which were added and which were removed. The payoff is a pairing that survives insertion: add a paragraph at the top of B and everything below it still lines up against A, instead of every remaining line reporting as changed.
The result is four columns wide, a line number and its content for each side. Removed lines sit tinted red on the left with the right cell blank, added lines sit green on the right with the left cell blank, and unchanged lines carry both numbers. A tally of added, removed and unchanged runs above the grid.
Comparison is line-level rather than word-level, so altering one character marks the whole line removed and re-added instead of highlighting the character itself. Both versions stay in the page, which matters when the thing being compared is a contract or a production config.
3 added, 1 removed, 146 unchanged.# and Original (A) stay visible through a long file.The table holds one number per pair of lines, so cost grows with the product of the two line counts rather than their sum. Two 500-line files is 250,000 cells and returns at once. Two 2,000-line files is 4 million cells, still workable but visibly slower. Two 10,000-line files is 100 million cells and will lock the tab up, so split the file or reach for a command-line diff at that size. Nothing is uploaded and no ceiling is enforced, which means your browser is the only thing standing between you and that 100-million-cell table.