Brute Force Calculator
Password strength meters lie by omission: they check whether you used a symbol without computing what that symbol actually bought you. The honest metric is entropy, borrowed from Claude Shannon’s information theory: E = L x log2(R), where L is length and R is the size of the character pool. Every bit doubles the attacker’s work. A 12-character password drawn from all four character classes carries 78.7 bits; an 8-character lowercase one carries 37.6. That gap is the difference between seconds and millennia.
The Brute Force Calculator below computes entropy, total combinations, and estimated crack time across four attack speeds, from a throttled login form to an offline GPU rig. Note what it does not ask for: your password. Length and character classes are all the math needs, so there is nothing sensitive to leak. The worked example matches the tool exactly.
This tool never asks for your password: length and character classes are all the math needs.
Quick Answer
Key takeaways
- Entropy in bits is the real strength metric: each bit doubles the attacker's work.
- Length beats complexity: adding one character adds log2(pool) bits, roughly 6.6 for a full pool.
- Crack time depends on attack speed: a throttled login and a leaked hash differ by a factor of billions.
- 60 bits is the practical floor for important accounts; 80+ resists offline GPU attacks.
- Quick Answer
- Entropy: the one number that measures password strength
- From bits to years: how crack time is estimated
- The four attack speeds (and why one number lies)
- Worked example, checked against this exact tool
- What competing password calculators miss
- Frequently asked questions
- Add This Calculator to Your Website
- References
Entropy: the one number that measures password strength
Shannon entropy for passwords is simple: E = L x log2(R), with L the length and R the pool size. The pool is not fixed: all-lowercase gives R=26, adding uppercase makes 52, digits make 62, symbols make 94. This is the step naive strength meters skip: they check that a symbol is present without recomputing the keyspace it unlocks. Total combinations are R^L, and the expected crack time is half the keyspace divided by the guess rate (the attacker finds it halfway through, on average).
| Password shape | Pool | Entropy | GPU-rig crack time |
|---|---|---|---|
| 8 chars, lowercase | 26 | 37.6 bits | ~10 seconds |
| 8 chars, all classes | 94 | 52.4 bits | ~3.5 days |
| 12 chars, all classes | 94 | 78.7 bits | ~754,000 years |
| 16 chars, all classes | 94 | 104.9 bits | ~59 trillion years |
From bits to years: how crack time is estimated
The pipeline is: combinations = R^L, average guesses = combinations / 2, seconds = average guesses / guesses-per-second, then convert to a human unit. The strength bands the calculator uses: under 28 bits very weak, 28-35 weak, 36-59 reasonable, 60-127 strong, 128+ very strong. Note the honest caveat baked into the results note: entropy measures brute-force resistance only. A 60-bit password built from dictionary words falls to a wordlist attack far faster than the math suggests.
The four attack speeds (and why one number lies)
A single "time to crack" figure is meaningless without the attack speed, so the calculator models four: a throttled online login (~100/hour), an unthrottled online form (1,000/sec), an offline slow hash like bcrypt (10,000/sec), and an offline fast hash on a GPU rig (10 billion/sec). The same 12-character password that survives a throttled login for geological time falls to a GPU rig in 754,000 years, and an 8-character lowercase password survives the login form for centuries but dies in seconds offline. Showing the spread, not just the flattering number, is the point: the threat model is the hash leaking, not the login form.
Worked example, checked against this exact tool
Length 12, all four character classes checked (pool = 94), offline GPU rig at 10 billion guesses/sec: entropy = 12 x log2(94) = 78.7 bits; combinations = 94^12 = 4.76 x 10^23; average guesses = 2.38 x 10^23; seconds = 2.38 x 10^13; crack time = about 754,000 years, verdict "Strong". Enter the defaults into the calculator and every figure matches.
What competing password calculators miss
Most password strength tools demand your actual password, transmitting the most sensitive string you own to somebody's server for analysis. This calculator's first design decision is privacy: length and character classes are sufficient for the entropy math, so the password field does not exist. The second gap is single-speed crack times: competitors that show one number usually pick the flattering online rate. The third is the missing caveat: entropy assumes random selection, and dictionary-based passwords need a wordlist warning, not a bit count. This tool shows all four attack speeds and states the dictionary caveat under every result.
Frequently asked questions
The unpredictability of a password measured in bits: entropy = length x log2(character pool size). Each bit doubles the guesses an attacker needs. 60 bits is strong for general use; 80+ resists offline GPU attacks.
60 bits is the practical floor for important accounts; 80+ bits withstands offline GPU brute force. Under 28 bits is very weak and under 36 weak: change those passwords first.
Usually yes. Length adds log2(pool) bits per character, so four random words (28 lowercase characters plus spaces) reach about 133 bits: far beyond a short complex password, and easier to remember. NIST treats passphrases as a preferred form of memorized secret.
Because it does not need it. Entropy math needs only length and character classes, so there is nothing sensitive to transmit or leak. Any strength tool that demands your actual password is taking an unnecessary risk with it.
It ranges from 100 guesses per hour against a throttled login form to 10 billion per second against a leaked fast hash on a GPU rig. Slow hashes like bcrypt cut offline speeds to about 10,000/sec. The threat model that matters is the leak, not the login form.
No. NIST SP 800-63B prioritizes length over composition: no forced uppercase/number/symbol rules, no scheduled rotation without evidence of compromise, minimum 8 characters with 15+ recommended, and screening against known-breached passwords.
Yes, and that is the key caveat. Entropy assumes random selection from the pool. A password built from dictionary words or patterns falls to wordlist attacks far faster than its bit count suggests. Randomness or a password manager is what makes the math real.
Add This Calculator to Your Website
References
- Dev.to, "I Built a Password Strength Calculator That Never Sends Your Password Anywhere: Here's the Entropy Math" - E = L x log2(R), pool detection, crack-time tiers. Dev.to: Password Entropy Math. Accessed October 19, 2026.
- NumericCore, "Password Entropy and Strength Calculator" - entropy equation, 60-bit strong threshold, 80+ military-grade. NumericCore: Password Entropy Calculator. Accessed October 19, 2026.
- NeftCal, "Password Strength / Entropy Calculator" - four attack-speed tiers, strength bands, expected crack time formula. NeftCal: Password Strength Calculator. Accessed October 19, 2026.
- OnlyCalculators, "Password Entropy Calculator" - worked 12-character example, 71.5-bit walkthrough. OnlyCalculators: Password Entropy. Accessed October 19, 2026.
- Dev.to, "Building a Password Strength Analyzer with Entropy and Crack Time" - JavaScript analyzer, charset detection, strength labels. Dev.to: Password Strength Analyzer. Accessed October 19, 2026.
- NIST, "Conformance Criteria for NIST SP 800-63B" - memorized secret length requirements and assessment objectives. NIST: SP 800-63B Conformance. Accessed October 19, 2026.
- Bellator Cyber, "NIST Password Manager Guidance: SP 800-63B-4 (2025)" - memorized secret standard, no composition rules, no scheduled rotation. Bellator: NIST Password Guidance. Accessed October 19, 2026.
- LMG Security, "Finding the Optimal Minimum Password Length" - real cracking-rig data: 8-char NTLM under 8 hours, 12-char 77,000 years. LMG: Minimum Password Length Data. Accessed October 19, 2026.