How it works
- UUID v4 uses
crypto.randomUUID(), the browser's built-in generator for RFC-4122 version 4 UUIDs — collision probability is astronomically low, and it's the safest default. - UUID v7 puts a millisecond Unix timestamp at the front, so IDs generated later sort after earlier ones. That makes them ideal as database primary keys — you get uniqueness and time-ordering without a separate sort column.
- UUID v1 is the classic time-based UUID (timestamp + node identifier). UUID v6 is v1 with the time fields reordered so the value sorts chronologically, a drop-in upgrade where v1 was already in use.
- Nil UUID is the all-zeros special value (
00000000-0000-0000-0000-000000000000), sometimes used as a placeholder or "no ID" sentinel. - Short ID is an 8-character alphanumeric ID — more compact than a full UUID, at the cost of a higher (but still very low) collision chance, good for short-lived or low-volume use cases.
- Nano ID style mimics the popular nanoid library's format: 21 URL-safe characters, a common alternative to UUIDs in modern web apps.
Generate RFC-4122 UUIDs — version 4 (random) and version 1 (time-based) — singly or in bulk, with formatting options. Runs locally, so IDs are generated on your device.
Frequently asked questions
What's the difference between UUID v4 and v1?
v4 is randomly generated and reveals nothing about when or where it was made — the safest default. v1 is based on timestamp and a node identifier, which can be useful when you want roughly sortable IDs.
Are these UUIDs unique?
v4 UUIDs draw from 122 random bits, so the chance of a collision is astronomically small — unique enough for database keys, filenames and distributed systems in practice.
Can I generate many at once?
Yes — bulk mode produces a list you can copy in one go, handy for seeding test data or pre-generating keys.
Is generation cryptographically random?
v4 uses the browser's crypto random source where available, giving strong, unpredictable values suitable for identifiers.