In short: a UUID is a random 128-bit ID — unique without coordination. Generate free UUIDs →

The version-4 format

The most common flavour, UUID v4, is randomly generated with a fixed pattern:

xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx
8-4-4-4-12 hex digits · the 4 marks version 4 · y is 8, 9, a or b (the variant)

Other versions exist: v1 embeds a timestamp and MAC address, v3/v5 derive from a namespace plus a name (MD5/SHA-1), and v7 — increasingly popular — is time-ordered for database-friendly sorting.

Why collisions basically never happen

A v4 UUID carries 122 random bits, giving about 5.3 × 10³⁶ possible values. To grasp that: you could generate a billion UUIDs per second for a century and your chance of a single collision would still be vanishingly small. This is what makes UUIDs safe for distributed systems — thousands of servers can create IDs simultaneously without ever asking a central authority "is this one taken?"

Where UUIDs show up

  • 🗄️ Database primary keys — especially across merged or replicated databases
  • 🔑 Session IDs and API tokens — unguessable by design
  • 📦 Distributed systems — event IDs, message queues, microservices
  • 💾 Filesystems and partitions — your disk's partitions are very likely UUID-identified

❓ UUID FAQs

What does UUID stand for?

Universally Unique Identifier — a 128-bit value written as 32 hex digits in five hyphen-separated groups.

What does a UUID look like?

Like 550e8400-e29b-41d4-a716-446655440000. In v4, the third group's first digit is always 4.

Can two UUIDs be the same?

Theoretically yes, practically no — 122 random bits give ~5.3 × 10³⁶ possibilities, making collisions astronomically unlikely.

What are UUIDs used for?

Database keys, session IDs, distributed systems and APIs — anywhere unique IDs must be created independently.

Related: What Is a QR Code? · What Is Base64? · Free UUID Generator

UUIDs are defined by RFC 9562 (formerly RFC 4122). Never use predictable UUIDs for security tokens — use a cryptographic random source.