How is Levenshtein distance calculated?

It uses a dynamic programming matrix. Each cell dp[i][j] holds the edit distance between the first i characters of string A and the first j characters of string B. When characters match the cost carries over diagonally; otherwise the cell is 1 plus the minimum of the cell above (deletion), left (insertion), or diagonal (substitution). The bottom-right cell is the final distance.

What counts as one edit?

One insertion (add a character), one deletion (remove a character), or one substitution (replace one character with another). Transpositions (swapping adjacent characters) are NOT counted — use Damerau-Levenshtein for that.

What is the similarity percentage?

Similarity = (1 - distance / max(len1, len2)) * 100. Two identical strings score 100%. Completely different strings of the same length score 0%.

Is the calculation case-sensitive?

Yes by default. Toggle "Case-insensitive" to fold both strings to lowercase before computing the distance.

Why does the matrix have an extra row and column?

The first row and column represent the cost of transforming an empty string into each prefix of the other string — they are initialized to 0, 1, 2, ... and serve as the base cases for the recursion.

How large can the strings be?

The tool handles strings up to 60 characters each in the visual matrix view (3600 cells). For larger strings the distance is still computed correctly but the matrix is hidden to keep the page usable.

Levenshtein Distance Calculator

Name: Levenshtein Distance Calculator
Availability: InStock
Author: Nham Vu

Find the minimum number of edits needed to transform one string into another, with a full DP matrix and highlighted edit path.

Strings to Compare

Source String

Target String

Case-insensitive comparison

Enter two strings and click Compute to see the DP matrix.

Copied!

Summary

Find the minimum number of edits needed to transform one string into another, with a full DP matrix and highlighted edit path.

How it works

Enter the source string in the left input and the target string in the right input.
The tool builds an (m+1) x (n+1) dynamic programming matrix in real time.
Each cell stores the minimum edits needed to transform the substring up to that point.
The bottom-right cell is the final Levenshtein distance between the two strings.
The optimal edit path is highlighted in the matrix so you can trace each operation.
Similarity percentage is computed as 1 minus (distance / max string length).

Use cases

Tune fuzzy-search thresholds for autocomplete and spell-check features.
Measure string similarity between database records for deduplication.
Understand edit distance for NLP, DNA sequence alignment, and diff algorithms.
Debug or visualize the DP matrix during algorithm learning.
Compare usernames, product codes, or identifiers for near-duplicates.
Estimate how many keystrokes a user made when mis-typing a password.

Frequently Asked Questions

Last updated: 2026-06-13 · Reviewed by Nham Vu