Evaluate a password's entropy
Measure the entropy in bits of your password (its mathematical resistance to randomness) and estimate the time needed to crack it by brute force at different computing speeds. Nothing is sent anywhere: the calculation happens entirely in your browser.
How it works
Entropy measures, in bits, the average number of attempts needed to guess a password by pure brute force: it's calculated from the password's length and the size of the character set used (lowercase, uppercase, digits, symbols). The higher the entropy, the more the number of possible combinations grows exponentially, and the longer the estimated cracking time becomes. The table above applies this entropy to several realistic computing speeds, from a limited online attack to a dedicated GPU cluster for cracking hashes.
Does the entropy estimate guarantee that a password is safe?
No. This calculation assumes a pure brute-force attack, which tests all possible combinations at random. In practice, attackers use dictionaries of leaked passwords and common transformation rules (capital letter at the start, digit at the end, substitutions "a" → "@") that crack a password based on a dictionary word much faster, even if it's long. A random password generated by a manager is always preferable to a memorable phrase, even if its theoretical entropy is similar.
Why does the cracking time vary so much depending on the scenario?
Computing speed depends heavily on context: an online service that limits the number of attempts (often a few per second) offers very different protection from a stolen hash then cracked offline on dedicated hardware, capable of testing tens of billions of combinations per second. This is why securely storing passwords (with slow hashing functions like bcrypt or argon2) is just as important as the choice of password itself.
Privacy: the entered password is never sent over the network — the entropy calculation and the cracking time estimate run entirely in your browser, locally, and are neither stored nor transmitted.