b2KIT

Diceware Passphrase Generator

Generate strong passphrases using the Diceware method with EFF word lists. Estimate entropy and crack time.

Tested tool guide Tested browser tools Checked August 16, 2026

What Diceware Passphrase Generator does and how it behaves

Diceware Passphrase Generator creates a passphrase by independently selecting words from an EFF Diceware list, then reports its entropy and an estimated time to search the possible combinations. Adding words increases the search space exponentially while keeping the result relatively readable. The main trap is replacing generated words with preferred ones while trusting the original estimate: that estimate assumes random word selection. Generation occurs in the browser, and the passphrase is not uploaded.

How the result is produced

1

Word selection

The EFF long list has 7,776 entries, corresponding to the 6^5 possible outcomes of rolling five six-sided dice. Each outcome identifies one word, and successive selections form an ordered passphrase. The browser tool performs the selections rather than requiring manual dice rolls. Spaces or other separators improve readability but are not the source of the word-selection entropy.

2

Strength calculation

For N equally likely words and k independent selections, the possible sequence count is N^k and the entropy is k times log2(N). With the 7,776-entry EFF long list, each word contributes about 12.925 bits, so six words provide about 77.55 bits. The crack-time result converts this search space using a guessing-rate model; it is an estimate, not a guarantee.

Good uses

  • Creating a memorable master passphrase for a password manager that accepts long passwords.
  • Generating a word-based secret for full-disk encryption or an encrypted archive.
  • Comparing how different Diceware word counts affect entropy and estimated cracking time before setting a password policy.

Limits and checks

  • The entropy calculation assumes each word is selected independently and uniformly. Themes, favorite words, or manual substitutions introduce bias that the displayed number does not represent.
  • Crack time depends on the assumed guessing rate. Online rate limits, offline cracking hardware, password hashing, and attacker knowledge can produce very different conditions.
  • A generated phrase may violate a service's maximum length or allowed-character rules. Reusing it also lets one compromise expose every account sharing the phrase.

Common questions

How many Diceware words should I generate?

There is no universal count. Match the word count to the consequence of compromise and the destination's length rules. On the 7,776-entry EFF long list, one independent word contributes about 12.925 bits, so six words represent about 77.55 bits before any human-chosen edits. More independently selected words increase the entropy linearly.

Will the same settings reproduce the same passphrase?

No. The selected list and word count determine the form and theoretical entropy, not a reproducible phrase. Each generation is intended to make new word selections. Store the chosen result securely if you need it again; the settings alone are not enough to reconstruct it.

References and verification

The behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.

Related Tools