Frequency distribution

Build a character, byte, or n-gram frequency table from text or hex — counts and percentages, ranked most to least frequent. Runs in your browser; nothing is uploaded.

About this tool

Frequency distribution counts how often each symbol appears in your input and ranks the symbols from most to least frequent, with both a raw count and a percentage of the total. It's the building block of classic frequency analysis.

Choose what to tally:

Characters (default): each Unicode character. Spaces, tabs and newlines are shown with readable labels (␠ (space), \t (tab), \n (newline)).
Bytes: the raw bytes (shown as 0xNN). A character like é is two UTF-8 bytes, so byte and character counts differ for non-ASCII text.
N-grams: overlapping windows of n characters (set N-gram size — 2 gives bigrams, 3 trigrams, and so on).

You can read the input as plain text or as a hex string (whitespace and an optional 0x prefix are ignored) — handy for analyzing binary or encoded data.

Privacy

Everything runs in your browser via WebAssembly — your input is never uploaded to a server. You can also run it from the gizza CLI or inside a gizza chat (which return the distribution as structured JSON with counts and percentages).

Common uses

Classic letter-frequency analysis for cryptograms and substitution ciphers.
Inspect the byte histogram of a file or payload (paste it as hex).
Find the most common bigrams / trigrams in a body of text.
Spot skew or padding characters in encoded data.

FAQ

Can I count 'A' and 'a' as the same letter?

Yes — turn case sensitive off and characters and n-grams are folded to lowercase before tallying, so The and the land in the same bucket. Byte mode ignores the switch: raw bytes have no case, so 0x41 and 0x61 always stay separate there.

Do the n-grams overlap?

Yes. N-gram mode slides a window one character at a time, so banana with size 2 yields ba, an, na, an, na — an and na each count twice. The size must be at least 1; size 2 gives the classic bigram table used in cipher analysis.

How do I analyze binary data instead of text?

Set the input kind to hex and paste the bytes as a hex string — whitespace and a leading 0x are ignored. An odd number of hex digits or a non-hex character is reported as an error rather than silently skipped, so you know the paste was clean.

Why don't byte counts match character counts for my text?

UTF-8. ASCII letters are one byte each, but é is two bytes and most emoji are four, so byte mode splits one character across several 0xNN entries. Use char mode for human-readable text and byte mode when you care about the raw encoding.

Developer & Automation Access

Run it from the terminal

Same engine as this page, headless — via the gizza CLI:

gizza tool frequency-distribution "Paste text to analyze…"

New to the CLI? Get gizza →

Open it by URL

Pre-fill and auto-run this tool with query parameters — the names match the API/CLI:

https://gizza.ai/tools/frequency-distribution/?input=Paste%20text%20to%20analyze%E2%80%A6&input_kind=text&mode=char&n=2&case_sensitive=true

Machine-readable descriptor: tool.json — title + parameters JSON Schema for agents.