UUID Generator

What Is a UUID?

A UUID (Universally Unique Identifier) is a 128-bit identifier standardized by RFC 4122. It is written as 32 hexadecimal digits displayed in five groups separated by hyphens, following the pattern 8-4-4-4-12 (e.g., 550e8400-e29b-41d4-a716-446655440000). UUIDs are designed to be globally unique without requiring a central registration authority, which makes them indispensable in distributed computing where multiple systems need to independently generate identifiers that will never collide.

The RFC 4122 standard defines five versions of UUIDs. Version 1 uses the current timestamp and the machine's MAC address. Version 3 and Version 5 generate deterministic UUIDs from a namespace and a name using MD5 and SHA-1 hashing respectively. Version 4, which this tool generates, uses cryptographic randomness for all significant bits. Version 4 is by far the most widely used in modern software because it requires no coordination between generators and exposes no information about the generating machine.

This generator uses your browser's crypto.randomUUID() method, which is part of the Web Crypto API. This API uses a cryptographically secure pseudo-random number generator (CSPRNG), providing the same quality of randomness used for encryption keys. No data is transmitted to any server.

How UUID v4 Generation Works

A UUID v4 is 128 bits long, but not all bits are random. Four bits are reserved for the version number (0100 for v4, placed in the most significant nibble of the 7th byte), and two bits are reserved for the variant indicator (10 in the two most significant bits of the 9th byte). This leaves 122 bits of randomness, producing approximately 5.3 x 10³⁶ possible unique values.

To put that number in perspective: if you generated one billion UUIDs per second, it would take about 85 years before the probability of a single collision reached 50%. In practical terms, UUID v4 collisions are effectively impossible for any real-world application. The crypto.randomUUID() function handles the bit-level formatting automatically, producing a properly structured UUID string in a single call.

Common Use Cases

• Database primary keys — UUIDs are used as primary keys in databases like PostgreSQL (which has a native UUID type), MySQL, and MongoDB. Unlike auto-incrementing integers, UUIDs can be generated on any node without coordination, which is critical for distributed databases and microservices.
• API request and correlation IDs — Assigning a UUID to each API request enables end-to-end tracing across microservices. Log aggregation tools can filter by the correlation ID to reconstruct the full request path.
• Session tokens — Web applications use UUIDs as session identifiers. The cryptographic randomness of v4 UUIDs makes them unpredictable, which is important for security.
• File and resource naming — Cloud storage services often use UUIDs to name uploaded files, preventing naming collisions when multiple users upload files simultaneously.
• Message queue deduplication — Message brokers like RabbitMQ, Kafka, and SQS use UUIDs as message IDs. Consumers can use these IDs to detect and discard duplicate deliveries.
• Distributed system node identification — Each node in a cluster can generate its own UUID at startup to identify itself without consulting a registry.
• Test data generation — When writing integration tests, developers need unique identifiers for test records. Bulk-generating UUIDs provides ready-to-use test data.
• Offline-first applications — Apps that work offline (mobile apps, PWAs) can generate UUIDs for new records locally, then sync them to the server later without worrying about ID conflicts.

Tips and Best Practices

• Use lowercase by default. RFC 4122 specifies that UUIDs should be output in lowercase, though parsers must accept both cases. The uppercase option in this tool is provided for systems that require it.
• Store UUIDs as native types when possible. PostgreSQL's UUID column type stores UUIDs as 16 bytes, which is more efficient than storing the 36-character string representation. Most ORMs handle the conversion automatically.
• Consider UUID v7 for time-ordered needs. UUID v7 (defined in RFC 9562, published 2024) encodes a Unix timestamp in the most significant bits, making UUIDs sortable by creation time. This improves database index performance compared to fully random v4 UUIDs. However, v4 remains the standard choice when ordering is not needed.
• Do not rely on UUIDs for security tokens alone. While v4 UUIDs are cryptographically random, they are typically transmitted and stored in plaintext. For authentication tokens, combine UUIDs with cryptographic signing (e.g., JWTs) or use dedicated token generation libraries.
• Bulk generate for batch operations. Use the count input to generate up to 100 UUIDs at once and copy them all with a single click, which is much faster than generating them one at a time in code.

UUID vs Alternatives

Auto-incrementing integers: Simple and space-efficient (4 or 8 bytes vs 16 bytes for a UUID), but they require a central authority (the database) to assign the next value. This creates a bottleneck in distributed systems and leaks information about record counts and creation order.

NanoID: A popular JavaScript library that generates shorter, URL-safe random IDs (21 characters by default vs 36 for UUIDs). NanoID is a good choice for client-facing identifiers where URL length matters, but UUIDs have broader ecosystem support and a formal specification.

ULID (Universally Unique Lexicographically Sortable Identifier): ULIDs encode a timestamp followed by randomness, producing IDs that sort chronologically. They serve a similar purpose to UUID v7 but predate it and use a different encoding (Crockford Base32 instead of hex). ULIDs are 26 characters, shorter than UUIDs, but less universally supported.

Snowflake IDs: Used by Twitter and Discord, Snowflake IDs are 64-bit integers that encode a timestamp, machine ID, and sequence number. They are very space-efficient and time-sortable, but require centralized machine ID assignment, which makes them less suitable for truly distributed generation.