What Is a UUID? A Plain-English Guide to Unique Identifiers

If you have ever seen a string like f47ac10b-58cc-4372-a567-0e02b2c3d479 in a database, a log file, or an API response, you have met a UUID. They look intimidating, but the idea behind them is straightforward and genuinely useful. This guide explains what a UUID is, why two of them practically never clash, and where you will run into them, written for developers but readable by anyone.

What a UUID actually is

UUID stands for Universally Unique Identifier. Microsoft’s ecosystem often calls the same thing a GUID (Globally Unique Identifier), and the two terms are interchangeable.

A UUID is a 128-bit value, which is just a very large number. It is written as 32 hexadecimal digits split into five groups separated by hyphens, in an 8-4-4-4-12 pattern:

xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx

That is all it is: a fixed-length label whose only job is to be different from every other label. It carries no meaning by itself, which is exactly the point. It is a name you can hand out without coordinating with anyone.

Why collisions are practically impossible

The reason UUIDs work is the sheer size of the space they live in. A 128-bit value has roughly 3.4 times 10 to the 38th power possible combinations. That is a number with 39 digits in front of the decimal point.

When you generate a version 4 UUID, almost all of those bits are filled with random data. To get a duplicate, you would have to randomly produce the same value twice out of that astronomically large pool. The odds are so small that, in practice, you can treat every freshly generated UUID as unique without checking.

This is what makes them so handy: any machine, anywhere, can mint a UUID on its own and trust it will not collide with one created somewhere else, with no central authority handing out numbers.

Where UUIDs are used

Because they need no coordination, UUIDs show up wherever independent parts of a system have to name things without talking to each other.

• Database primary keys. Instead of a sequential id that reveals row counts and forces a central counter, a table can use a UUID, letting records be created on different servers at once.
• API request and trace ids. Tagging each request with a UUID makes it easy to follow a single call through logs across many services.
• Distributed systems. When data is created offline or on multiple nodes that later sync, UUIDs prevent two records from claiming the same id.
• File and resource names. Uploads, sessions, and temporary objects often get a UUID so their names never collide.

Version 4: the random one

UUIDs come in several versions, but the one you will use most is version 4, the random variant.

A v4 UUID fills almost all of its bits with random data, with a few reserved to mark it as version 4. The result needs no input, no timestamp, and no hardware address. You ask for one and you get a fresh random identifier. Its simplicity and lack of dependencies are why v4 has become the default choice for application code that just needs a unique key.

Other versions exist for specific needs, such as time-based ordering, but if you are not sure which to use, v4 is almost always the right answer.

Generate UUIDs in your browser

When you need one (or a batch) you can produce them instantly. The free UUID Generator creates valid version 4 UUIDs on demand, ready to copy into your code, your database, or a test fixture.

Because the UUID Generator runs entirely in your browser, the values are generated on your own machine with no sign-up and nothing sent to a server. That keeps it fast and means the identifiers never leave your device until you use them.

Related tools

A few neighboring generators cover related jobs. When you need a human-usable secret rather than an identifier, the Password Generator builds strong random passwords to your length and character rules. For arbitrary tokens, API keys, or test data of a custom length, the Random String Generator gives you full control over the alphabet. And when you need a fixed fingerprint of some content rather than a random value, the Hash Generator produces checksums like SHA-256.

The short version

A UUID is a 128-bit unique identifier written as 32 hex digits in an 8-4-4-4-12 pattern, and its huge address space makes collisions practically impossible. Version 4 is the random, dependency-free default used for database keys, request ids, and distributed systems. When you need some, the UUID Generator produces them in your browser instantly.