UUID Generator
Random version 4 UUIDs from cryptographic randomness — one at a time or a thousand at once, formatted the way your system needs them.
How UUIDs are generated
Each UUID takes 122 bits from the browser's cryptographic random source and stamps in the version and variant bits required by the standard. That's enough randomness that collisions are a theoretical curiosity, not a practical concern — which is the whole point of UUIDs.
Set the count
From a single UUID to 1,000 at once, one per line.
Match your format
Uppercase for systems that expect it, no dashes for compact 32-character keys.
Copy or download
Copy the whole list to the clipboard or save it as a .txt file for seeding data.
Why collisions don't matter in practice
A v4 UUID has 2¹²² possible values — about 5.3 undecillion. Even generating a billion UUIDs per second for a century, the odds of a single collision remain negligible. That math is what lets independent systems mint IDs without ever talking to each other: microservices tagging requests, databases assigning primary keys, apps naming files. One caveat worth repeating: UUIDs are identifiers, not secrets. They're fine as IDs in URLs, but access control should never rest on an ID being hard to guess.
Frequently asked questions
What is a version 4 UUID?
A 128-bit identifier where 122 bits are random, formatted as 8-4-4-4-12 hex digits, like 550e8400-e29b-41d4-a716-446655440000. The "4" at the start of the third group marks the version.
Can two generated UUIDs collide?
In theory yes, in practice no: with 122 random bits, you'd need to generate about 100 trillion UUIDs to reach even a one-in-a-billion collision chance. That's why systems worldwide use them as unique IDs without coordination.
Are these UUIDs cryptographically random?
Yes — generation uses the browser's crypto API (crypto.randomUUID or crypto.getRandomValues), not Math.random. Still, UUIDs are identifiers, not secrets: don't use them as passwords or tokens with security guarantees.
What are UUIDs used for?
Database primary keys, API request IDs, file names that can't clash, session and correlation IDs in logs — anywhere you need an ID that's unique without asking a central authority.