JSON Diff & Compare Tool

A JSON diff is a comparison between two JSON documents that shows exactly what changed — which fields were added, which were removed, and which values were modified. It is the JSON equivalent of git diff, applied to structured data instead of source code.

JSON is widely used in API responses, configuration files, database exports, and application state. When a JSON payload changes — between API versions, environment configs, or test runs — a visual diff immediately answers the question “what exactly is different?” without manually scanning two documents.

This tool uses a line-level diff based on the Longest Common Subsequence (LCS) algorithm. Both inputs are first normalized to identical formatting (2-space indentation, sorted keys within each object), so purely cosmetic differences — whitespace, key order — are eliminated before the comparison runs.

The comparison uses the Longest Common Subsequence (LCS) algorithm — the same algorithm used by diff and git diff. It finds the largest set of lines that appear in both documents in the same order, then treats everything outside the LCS as either a deletion (from A) or an insertion (into B).

The algorithm pre-processes both JSONs through JSON.parse → JSON.stringify(…, null, 2) before diffing. This means a minified JSON and a pretty-printed version of the same data will compare as identical — the diff reflects structural and value changes, not formatting choices.

The side-by-side layout mirrors the classic git diff view. Each row in the diff table represents a line. Rows are aligned so that matching unchanged lines appear at the same vertical position, making it easy to see which changes are adjacent.

• Line-level, not semantic. The diff operates on formatted text lines, not JSON values. If a nested object moves position within the document, the diff will show it as deleted in one place and added in another rather than as a “move” operation.
• Key ordering. JSON.stringify preserves JavaScript insertion order. If the same keys appear in different orders in A and B, the formatted output will reflect that, and the diff will show changes even if all values are the same.
• Large documents. The LCS algorithm has O(m×n) complexity. Documents producing more than 2,000 lines each may take a second or two on older devices.
• No semantic array diff. Arrays are compared as sequences of lines. Inserting an element at position 0 of a 10-element array will look like “every line changed” rather than “one element inserted.”