Random Data Generator
Addresses, Dates And Test Tables

What A Random Data Generator Does For You

Real work needs fake data. A form needs something to submit, a table needs rows to sort, and a firewall rule needs an address to block. So a random data generator fills those gaps in seconds, without you inventing values by hand.

Pick a shape at the top, say how many you want, and press Generate. Then the values appear in the box below, ready to copy or to save.

Nine shapes share the same panel. As a result, one page covers addresses, IPs, dates, times, fractions and whole tables of records.

Everything runs in your browser. So nothing you make here reaches this site, and we store nothing.

How The Random Data Generator Works

First, the tool reads the boxes on screen. Then it asks your browser for random numbers. Next, it shapes those numbers into the form you asked for, whether that is an address, a date or a row of a table.

No value gets a head start. Because the tool throws away the numbers that would tip the odds, every address and every date keeps the same chance as its neighbours.

A Random IP Address Generator

IP addresses is the first shape, and it is the one most people arrive for. So choose IPv4 or IPv6, then pick the range you need.

Any address gives you the whole space. Public only skips the private and reserved blocks, while Private only stays inside them. Loopback keeps everything on 127.

Public, Private And Loopback Ranges

The private ranges are the ones your router hands out: 10, 172.16 to 172.31, and 192.168. They are set out in RFC 1918, and the tool follows that list exactly.

Public only goes the other way. It drops those three blocks, and it also drops loopback, multicast and the link-local 169.254 range. Therefore the addresses you get look like addresses from the open internet.

Addresses Inside Your Own Block

Testing a subnet works best when the addresses sit inside it. So type something like 192.168.4.0/24 in the block box, and every address lands in that range.

The tool rounds your entry down to the start of the block. For example, 192.168.5.77/24 becomes 192.168.5.0/24, which is the block you really meant.

Random IPv6 Addresses

Switch the version to IPv6 and the output changes shape. Short IPv6 form squeezes the longest run of zero groups into a double colon, which is how people write these addresses by hand.

Untick it and you get all eight groups in full. That longer form suits a config file, while the short form suits a document.

Which Addresses Are Safe To Publish

A random public address belongs to somebody. So keep it out of screenshots, blog posts and public tickets.

For documents, use the ranges set aside for exactly that job in RFC 5737: 192.0.2.0/24, 198.51.100.0/24 and 203.0.113.0/24. Type one of them into the block box, and the tool stays inside it. The full picture lives in the IANA special-purpose address registry.

A Random MAC Address Generator

MAC addresses is the second shape. So pick the separator your tooling expects: colons, hyphens, Cisco dots or nothing at all.

Capital letters are one tick away. Besides, the address type box lets you force the two bits that matter.

Locally Administered And Multicast Bits

The first byte of a MAC address carries two flags. One says the address is multicast, and the other says it belongs to the local network rather than to a hardware vendor.

Locally administered sets that second flag, so your fake address cannot collide with a real card. Multicast sets the first one instead. In short, the tool makes addresses that are honest about being made up.

Keep A Vendor Prefix

Sometimes a test needs addresses that look like one vendor. So type a prefix such as 00:1A:2B, and the tool keeps it while it randomises the rest.

Those first three bytes are the vendor block. The registry behind them sits in the IANA ethernet numbers list, so pick a prefix that suits the kit you are pretending to have.

A Random Date Generator

Dates is the third shape. So set the earliest and the latest date, then choose the way you want them written.

ISO gives you 2026-03-09, which is the format in ISO 8601 and the one databases prefer. US and European formats use slashes, and Unix seconds gives you a plain number for a timestamp column.

Weekdays Only

Business data rarely lands on a Sunday. So tick Weekdays only, and every date comes back between Monday and Friday.

That one tick makes a test invoice table far more believable. Also, it saves you filtering the weekends out later.

A Random Time Generator

Times is the fourth shape, and it works the same way. Set the earliest and the latest time on a 24 hour clock, such as 09:00 and 17:30.

Then choose the clock for the output. The 24 hour clock gives 14:05, while the 12 hour clock gives 2:05 PM. Seconds are optional, so a log line and a meeting slot both fit.

A Random Fraction Generator

Fractions is the fifth shape, and teachers use it most. Proper fractions stay below one, improper fractions sit above it, and mixed numbers put a whole number in front.

Set the largest denominator to match the class. For example, twelve suits early work, while ninety-nine suits a harder sheet. Lowest terms starts ticked, so untick it when you want the pupils to do the reducing.

A Random Test Data Generator For Tables

The last four shapes build a whole table of records. So tick the columns you need, set how many rows, and pick JSON, XML, CSV or TSV.

Fifteen columns are on offer: a row number, a UUID, a name, an email address, a city, a country, a date, a time, true or false, a whole number, a price, a status word, an IP address, a MAC address and a short sentence.

JSON, XML, CSV And TSV

JSON comes out as an array of objects, indented by default. Untick the indent box for a compact single line, which is what an API usually sends.

XML wraps the rows in tags you name yourself. CSV follows RFC 4180, so it quotes any field holding a comma, a quote or a line break. TSV is the same table with tabs, which pastes straight into a spreadsheet.

Names And Email Addresses That Are Safe To Use

The names are built from two short lists, so they read like people without belonging to anybody. Meanwhile the email addresses always end in example.com, example.net, example.org or test.example.

RFC 2606 keeps those domains for documentation. Therefore a test run can never post mail to a real stranger by mistake.

UUID Columns

Tick the UUID column and each row gets a version 4 identifier, with the version and variant bits set as RFC 9562 describes.

Those identifiers are random, not sequential. So they suit a primary key in a test database, and they never hint at how many records exist.

Repeatable Runs With A Seed

A seed is a short word that fixes the output. So type release-42 in the seed box, and the same settings always give the same values.

That matters for a test suite. First, commit the seed next to the test. Then anybody who runs it gets the same table, and a failure turns into something you can repeat.

Where The Randomness Comes From

Without a seed, the numbers come from the secure source built into your browser. For the details, see the MDN page on crypto.getRandomValues.

With a seed, the tool uses a small repeatable generator instead. Even so, the spread stays even. The only difference is that you can replay the run.

Copy, Download And Limits

Press Copy for the clipboard, or Download for a file. Each shape saves under a sensible name, so a JSON run lands as test-data.json and an address run lands as random-ip-addresses.txt.

A plain list returns up to a thousand values, while a table returns up to five hundred rows. Those limits keep the page quick on a phone, and the counters under the box tell you how many values you got and how many of them were different.

Safe By Design

The tool works in your browser. It sends nothing, and it stores nothing. So the seed you chose, the block you typed and the table you made all stay on your own device.