UUID Generator
Generate UUIDs (GUIDs) — random v4, time-ordered v7, name-based v5 and more — up to 1,000 at a time, and read the version and creation time of any UUID.
Runs in your browser. Nothing you add is uploaded.
What the UUID Generator does
A UUID is a 128-bit identifier you can generate independently on any machine with a vanishing probability of collision — no coordination, no central sequence, no database round trip. This generator follows RFC 9562, the 2024 standard that replaced RFC 4122. Version 4 is 122 bits of randomness; version 7 starts with the creation time, so IDs sort in the order they were made, which keeps database indexes fast; versions 5 and 3 turn a namespace and a name into the same UUID every time. Make up to 1,000 at once in the format your code needs, or paste any UUID to see which version it is and, for time-based ones, when it was created. Everything uses your browser's cryptographic random source and stays on your device.
How to generate a UUID
- Pick a version: 4 for general use, 7 for database primary keys, 5 to derive an ID from a name such as a domain or URL.
- Choose how many you need — one for a quick test, or up to 1,000 for seed data. For version 5, type one name per line instead.
- Choose the format: standard, no hyphens, braces or urn:uuid:, and list them one per line, comma-separated, as JSON or as SQL values.
- Copy one, copy them all, or download the list. To check an existing UUID, switch to Inspect a UUID and paste it.
The UUID Generator runs entirely in your browser — nothing you enter is uploaded, stored, or logged.
When to use it
Assigning primary keys before insert
Generating the id client-side lets you build an entire object graph with valid references before anything reaches the database, which simplifies offline-first and optimistic-update patterns considerably.
Correlating requests across services
A UUID attached to an incoming request and propagated through every downstream call turns a distributed trace into a single greppable string.
Naming uploaded files safely
Storing an upload under a UUID rather than its original filename removes an entire class of path traversal and collision problems in one step.
Good to know
- Version 4 UUIDs are random, so they scatter across a B-tree index. On very large tables this hurts insert performance — UUIDv7, which is time-ordered, is the modern answer.
- UUIDs are case-insensitive by specification but conventionally written in lowercase. Normalise before comparing as strings.
- Stored as text a UUID is 36 characters; stored as a native binary or uuid column it is 16 bytes. On large tables that difference is worth having.
- Do not treat a UUID as a secret. It is unguessable, but it is not an access control mechanism.
Frequently asked questions
What is a UUID v4?
Should I use UUID v4 or v7?
What is the chance two UUIDs collide?
What is the difference between UUID v1, v4, and v7?
Can I use UUIDs as database primary keys?
Are these UUIDs cryptographically secure?
Why do all version 4 UUIDs have a 4 in the same position?
Related tools
JSON Formatter
Format, validate and minify JSON with exact error locations.
Random Number Generator
Generate fair random numbers: ranges, no repeats, decimals, PINs.
Regex Builder
Build a regular expression from blocks, with live matches.
SQL Formatter
Format, beautify, and minify SQL queries for Postgres, MySQL, and SQLite.