UUID Generator

UUID Generator – Create Unique IDs for Your Projects

UUIDs (Universally Unique Identifiers) are 128-bit identifiers that are practically guaranteed to be unique worldwide. This tool generates UUID version 4 (random-based) using your browser's native crypto.randomUUID(), plus UUID version 7 (time-based and sortable thanks to its timestamp). No tracking, no signup. Your UUIDs are generated right in your browser.

Choose how many you want (1 to 50) and whether the hex letters should be uppercase or lowercase. One click generates them all, ready to copy. Perfect for developers who need quick test data.

Specification: what is a UUID?

A UUID is a 128-bit identifier standardized in RFC 4122. It is usually shown as 32 hexadecimal digits in five groups, in the pattern 8-4-4-4-12. The version sits in the first digit of the third group, the variant in the first digit of the fourth group:

• Version 4: random-based, xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx
• Version 7: time-based, xxxxxxxx-xxxx-7xxx-yxxx-xxxxxxxxxxxx

The y marks the variant and is always one of 8, 9, a or b (binary 10xx) for these versions.

UUID v4: random-based

UUID v4 uses 122 bits of real randomness (the remaining 6 bits are reserved for version and variant). The randomness comes from crypto.getRandomValues() or crypto.randomUUID(). The chance of a collision is so low you would have to generate billions of UUIDs before seeing a duplicate. For nearly every application, that is more than enough.

v4 example: 6f4d2c8a-1e3b-4a9f-9c21-8b7e5d4f3a10, random and therefore unordered.

UUID v7: time-ordered

UUID v7 starts with a 48-bit Unix timestamp in milliseconds, followed by version, variant and random data. This makes v7 UUIDs sortable by their creation time. That helps when they are used as primary keys in a database: newly inserted rows land in near-ascending order, the index stays compact and avoids frequent page splits. For data-heavy applications with large tables, v7 is often the better choice than fully random v4.

v7 example: 018f3c2a1e3b4a9f9c218b7e5d4f3a10, the leading part grows over time.

Where UUIDs are used

• Database primary keys globally unique IDs instead of auto-increment
• APIs resource IDs that can be generated client-side
• Session IDs temporary identifiers for user sessions
• File names avoid collisions when uploading files
• Microservices IDs that are unique across systems with no central coordination
• Test data generate dozens of unique IDs fast

Limits: what a UUID is not

A UUID is not a secret and not a security feature. It is only unique, not confidential; knowing an ID does not by itself grant access to protected data as long as access control relies on other checks. The uniqueness of v4 is statistical, not provable. And v7 sortability is only reliable within a single generator; across distributed systems with different clocks, the order may differ.

v4 or v7: which version when?

A simple rule of thumb helps. For lightweight applications where order does not matter, random v4 is enough. For database-heavy workloads where many new records are written one after another into an index, time-ordered v7 is often the better choice because it avoids fragmentation of index pages. Both are fully valid UUIDs; our generator creates both directly in the browser so you can choose per use case.

How many UUIDs until the first collision?

The collision probability for v4 grows only with the square root of the number of generated IDs (the so-called birthday paradox). As a rule of thumb, even with billions of generated UUIDs the chance of a duplicate stays vanishingly small. For normal databases, APIs and test data that is completely uncritical. Only if you generate astronomically many IDs and must rule out a duplicate, plan for additional checks or a central counter.

Frequently asked questions