UUID v4 Generator

About this UUID v4 Generator

The UUID v4 Generator creates random-style universally unique identifiers in the familiar 8-4-4-4-12 hexadecimal format. Modern browsers use crypto.randomUUID when available and the tool includes a cryptographic-random fallback.

A UUID (universally unique identifier) is a 128-bit value designed to be unique across systems and time without needing a central coordinating authority, making it the standard choice for database record IDs, API keys, and tracking identifiers in distributed software systems.

For related tasks, try Random Number Generator, Password Generator or QR Code Generator.

How to use it

1. Select Generate UUID v4.
2. Review the generated identifier.
3. Copy it into your application, test data or configuration.

Practical example

A UUID v4 looks like “550e8400-e29b-41d4-a716-446655440000”, with the version information represented in the appropriate position.

A typical version 4 UUID looks like f47ac10b-58cc-4372-a567-0e02b2c3d479, formatted as five groups of hexadecimal digits separated by hyphens, with the 4 in the third group specifically indicating this is a version 4 UUID.

Things to double-check

A version 4 UUID is generated using random or pseudo-random numbers, and while a collision is theoretically possible, the probability is so astronomically small under proper random generation that it's treated as effectively impossible for practical software engineering purposes.

What a version 4 UUID is

A UUID is a 128 bit identifier written as 32 hexadecimal digits in five groups, such as 123e4567-e89b-42d3-a456-426614174000. In version 4, 122 of the bits are random. The third group starts with 4, marking the version, and the fourth group starts with 8, 9, a or b, marking the variant. The tool creates them with the browser's secure generator, and you can generate as many as you like.

How unlikely a collision is

With 122 random bits there are about 5.3 times ten to the power of 36 possible values. You would need to generate roughly 2.7 quintillion UUIDs before there was even a 50 percent chance of one duplicate. That is why systems can create identifiers independently, without checking a central list. Collisions are possible in theory but far less likely than hardware failure, as long as the random source is good.

Using them in databases

Random UUIDs are popular as primary keys because they can be created anywhere without coordination. They are bigger than integers, 16 bytes in binary or 36 characters as text, and because they are random they scatter inserts across an index, which can slow large tables. Newer versions such as version 7 put a timestamp first so identifiers sort by creation time. Store them in a native UUID type where your database has one.

Not a secret

A UUID is an identifier, not a password. Do not use one alone as the only protection for a private link or a reset token, since people may log, share or guess patterns in how they are used. For secrets use a purpose-built token with enough randomness and an expiry. UUIDs are case-insensitive, so compare them in lowercase, and keep the hyphens consistent in your system.

Formats and storage

A UUID is shown as 36 characters with hyphens, but it can also be stored without them as 32 hexadecimal digits, or as 16 raw bytes, which is the most compact. Some systems wrap it in braces, and a URN form starts with urn:uuid:. Lowercase is the usual convention, though uppercase is equivalent. Pick one format for your system and convert at the edges. Using the native type of your database saves space and speeds up comparisons.

Where generated identifiers are used

UUIDs are used for order numbers, file names for uploads, session records, correlation identifiers in logs and idempotency keys in payment APIs, which let a server recognise a repeated request. They are long, so some systems shorten them by encoding the 16 bytes in a compact text form, which gives about 22 characters. If humans need to read or type the identifier, a shorter code with a check digit is easier. For machine use, a UUID is a convenient default because no central counter is needed.

Worked example: reading a UUID

Look at the identifier 550e8400-e29b-41d4-a716-446655440000. It has five groups of 8, 4, 4, 4 and 12 hexadecimal digits. The first digit of the third group is 4, which marks version 4, a randomly generated UUID. The first digit of the fourth group is a, which falls in the range 8, 9, a and b, marking the standard variant. The remaining digits are random. When you generate UUIDs with the tool, you can check these two positions to confirm that they follow the standard format.

How rare is a duplicate?

A version 4 UUID has 122 random bits. If you generated one billion identifiers every second, it would take about 86 years to reach even a 50 percent chance of a single duplicate. In practice, other failures are far more likely than a collision. The statement assumes a good random source: a poor generator, or one that is cloned across virtual machines with the same seed, can break the guarantee. The tool uses the browser's secure generator, which is designed to avoid that problem.

UUIDs as database keys

Random UUIDs are convenient as keys because any server can create them without asking a central counter. They have costs: they take 16 bytes instead of 4 or 8, and because they are random, new rows land all over an index, which can slow inserts and use more memory in very large tables. Time-ordered identifiers, such as UUID version 7 and similar formats, put a timestamp at the start so new rows are added near the end of the index. Choose based on the size of your system and the way that you query the data.

Practical uses and privacy

Beyond databases, UUIDs name uploaded files, tag log entries so that you can follow a request across services, and serve as idempotency keys that let a server recognise a retried request. Older UUID versions embedded the computer's network address and a timestamp, which could reveal information, and version 4 avoids that by being random. Validate incoming identifiers with a pattern or a library, store them in lowercase, and do not treat them as secrets, since an identifier that is easy to see in a link is not a password.

Key terms explained

A UUID is a 128-bit identifier written in five groups of hexadecimal digits. Version 4 is generated randomly. Version 7 starts with a timestamp so values sort by time. A GUID is the term used in Microsoft systems. A primary key uniquely identifies a database row. A collision is two identical identifiers. An idempotency key lets a server recognise a repeated request. A namespace UUID derives an identifier from a name. Entropy is the amount of randomness.

Rules of thumb and quick checks

A version 4 UUID has the digit 4 at the start of the third group and 8, 9, a or b at the start of the fourth. It has 122 random bits, so a collision is practically impossible. Store it in lowercase and in the native type of your database. Use time-ordered identifiers for large tables with many inserts. Do not use a UUID as a secret or a password.

Frequently asked questions