Unique identifiers without a central authority
Generate random UUID version 4 identifiers using your browser's cryptographic random source. Bulk generation up to 50 at once.
A UUID lets two systems that have never communicated both create identifiers without ever colliding. That is why they show up in databases, message queues and distributed logs.
Version 4 is almost entirely random — 122 bits of it. The odds of a collision are so remote that you would need to generate billions per second for a century to make one likely.
Fresh UUID v4 values, up to 50 per run Standard 8-4-4-4-12 hexadecimal formatting Optional uppercase, for systems that expect it Uses crypto.randomUUID where the browser provides it. The version digit is always 4 and the variant digit is 8, 9, a or b — that is the specification, not a quirk. Could two UUIDs ever be the same? In theory yes, in practice no. With 2¹²² possibilities you would need to generate a billion per second for about 85 years to reach a 50% chance of one collision.
Should I use a UUID as a database primary key? It is excellent for distributed systems where several nodes create records. The trade-off is index locality — random values scatter writes, which is why UUID v7 with a time prefix exists.
What is the difference between UUID and GUID? None technically. GUID is Microsoft's name for the same 128-bit standard.
Are these safe to use as security tokens? v4 uses a cryptographic random source, so they are unguessable. Even so, dedicated session tokens are the better tool for authentication.
Worked example Anatomy of a v4
Input: One generated identifier
Output: f47ac10b-58cc-4372-a567-0e02b2c3d479
Note: The 4 opening the third group marks the version; the a starting the fourth marks the variant. Everything else is random.
