Generate realistic synthetic datasets that preserve statistical properties without containing real personal information.
This tool builds tabular data from scratch. You declare the shape you want - column types such as name, email, integer, date, or city, plus a row count - and it returns a file of records that look like real data but correspond to no real person. Each column is sampled from its own value pool or range, so the output keeps the structure and rough distribution you asked for. The thing people most often miss: the output is random, so every run differs, and with few rows the observed distribution drifts from the target.
A concrete input and expected output from the current implementation.
Input
Columns: id (integer, sequential from 1), first_name (name), age (integer 18-99), city (US city). Rows: 10
Expected output
id,first_name,age,city 1,Emma,34,Austin 2,Liam,52,Denver 3,Sofia,27,Portland 4,Noah,41,Seattle 5,Mia,63,Boston 6,Ethan,19,Chicago 7,Ava,38,Nashville 8,Lucas,45,Phoenix 9,Zoe,29,Atlanta 10,Oliver,57,Columbus
The deterministic parts are exact: 10 rows, headers in the order requested, id running 1 through 10, and every age inside 18-99. Names, ages, and cities are sampled per run, so the values shown are one possible draw - a fresh run returns different values with the same shape and bounds.
You define the table column by column, not record by record. Each column gets a value type - person name, email address, integer with bounds, date, city, category - and a row count sets how many values each generator produces. Cells are then filled by sampling from the corresponding generator, so every column is internally consistent with its type but independent of the others.
The row count is honored exactly, and numeric, date, and category constraints are enforced per cell. The values themselves are random draws, so two runs of the same configuration produce different files with the same structure and bounds. If the tool offers a seed control, setting one makes the draw reproducible; otherwise, save the file you generated if you need stable rows.
Rows correspond to no real person, which is why this tool suits public demos and bug reports. It is not a privacy guarantee, though: name pools can still collide with living people, and describing sensitive real data in detail can leak facts through the description itself. For regulated data, treat synthetic output as a complement to, not a substitute for, proper de-identification.
Because the tool samples randomly rather than balancing quotas. With thousands of rows, the observed mix converges toward the target distribution; with a few dozen or hundred rows, small-sample noise is large, and a 5% category may appear at 2% or 8%. Ask for more rows, or fix exact counts per category if the tool supports it, rather than expecting small draws to be exact.
The example and behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.
Generate strong, random passwords with customizable length, character sets, and complexity requirements.
Analyze password strength with entropy scoring, crack-time estimates, and improvement suggestions.
Create cryptographically random strings for API keys, tokens, salts, and unique identifiers.
Create QR codes for URLs, text, Wi-Fi credentials, vCards, and more with customizable size and colors.
Scan and decode QR codes from uploaded images or your device camera to extract embedded data.
Compute MD5, SHA-1, SHA-256, and SHA-512 checksums to verify file integrity and detect tampering.