Password Strength Checker

Estimate real password entropy, detect patterns that make it weaker than it looks, and see honest crack-time estimates for different attack scenarios.

Calculator Security & Crypto Runs in your browser
Analysed entirely in your browser and never transmitted.
Try:

Result

Effective entropy
73.4 bits
Verdict
strong
Length
13 characters
Character pool
95 symbols
Naive entropy
85.4 bits
12 bits deducted for patterns

Crack time estimates

Attack scenarioGuess rateExpected time to crack
Online, rate limited (100/s)1e+2/s1.99e+12 years
Online, unthrottled (10k/s)1e+4/s1.99e+10 years
Offline, bcrypt (100k/s)1e+5/s1.99e+9 years
Offline, fast hash on GPUs (10^11/s)1e+11/s1989.3 years
Nation-state offline (10^14/s)1e+14/s2 years

Weaknesses detected

  • Contains a four-digit year - a standard mutation rule in cracking tools.
What actually helps
Length beats complexity. Four random words are stronger and far more memorable than eight characters of mangled substitutions - and cracking rulesets specifically target the substitutions.
Habit warning
Analysed entirely in your browser. Never type a real production password into any online checker, including this one.

About Password Strength Checker

Password strength is measured in entropy - how many guesses an attacker expects to need. Character-class rules are a poor proxy for it, because humans satisfy them in predictable ways that cracking tools model explicitly.

Why breach corpora beat brute force

Real attacks rarely start with brute force. They start with the billions of passwords already exposed in breaches, then apply mutation rules — capitalise, append a year, substitute a for @ — to each one. A password with 60 bits of theoretical entropy is worthless if it appears in a wordlist, and no strength meter can see that. Checking a password against a breach corpus (using a k-anonymity API so the password itself is never sent) tells you more than any entropy calculation, and should be part of any password policy that intends to work.

Why pattern penalties matter

Summer2026! scores well on naive entropy: eleven characters, four character classes, about 72 bits. In reality it matches a rule every cracking wordlist applies - capitalised dictionary word, year, symbol - and falls in seconds. Real strength comes from unpredictability, which is why generated passwords and long passphrases outperform anything a person composes.

Common use cases

  • Explaining to a team why the corporate password policy is not producing strong passwords.
  • Sanity-checking a passphrase scheme before adopting it.
  • Comparing the value of extra length against extra character classes.

Edge cases and gotchas

  • Estimates assume brute force; a password already in a breach corpus is cracked instantly regardless of entropy.
  • Never enter a live credential into any web tool, this one included.

Frequently asked questions

Should users be forced to change passwords regularly?
No. NIST SP 800-63B dropped that guidance because forced rotation drives predictable increments such as Summer2025 becoming Summer2026. Rotate on suspicion of compromise instead, and check against breach corpora.
How many bits is enough?
60 bits resists online attack comfortably. For anything protecting infrastructure, aim for 80+, and use a generated secret rather than a chosen one.