UUID generator
Generate RFC 4122 UUID v4 values instantly in your browser — cryptographically random every time. Batch-generate dozens at once for database seeding and API testing.
What is a UUID?
A UUID (Universally Unique Identifier) is a 128-bit identifier that's practically guaranteed to be unique without requiring a central registry. Version 4 UUIDs are randomly generated, producing 2^122 possible values — the probability of a collision is astronomically small.
UUIDs are widely used as database primary keys, API request identifiers, session tokens, and distributed system identifiers. The standard format is 8-4-4-4-12 hexadecimal characters separated by hyphens.
When should you use a UUID?
Database primary keys (often preferred over auto-increment in distributed systems to avoid coordination overhead), API request tracking for log correlation, session tokens, idempotency keys to make retries safe, unique file names, and identifiers in event streams.
UUID versions explained
v1 is timestamp + MAC address (fast but leaks identifying info). v4 is fully random (most common, what this tool generates). v5 is name-based with SHA-1 (deterministic for the same input). v7 is timestamp-sortable (newest, gaining adoption for database keys). This tool generates v4 UUIDs using the Web Crypto API for cryptographically strong randomness.
Need cryptographic hashes instead? Try our hash generator.
Common Uses
- Database primary key generation: Generate UUID v4 primary keys for distributed databases where auto-increment integers cause conflicts across nodes.
- Session and token creation: Produce cryptographically random session tokens for web applications where guessable IDs create security risks.
- File and asset naming: Name uploaded files with UUIDs to prevent collisions, avoid path traversal attacks, and enable easy deduplication.
- Microservice correlation IDs: Attach a UUID to every request as a correlation ID to trace it across logs in distributed microservice architectures.
- Test data generation: Generate realistic UUID records for seeding development databases, unit tests, and API mock responses.
- IoT device registration: Assign unique UUIDs to IoT devices at provisioning time to enable device-level telemetry tracking and management.
- Idempotency keys for payments: Generate UUIDs as idempotency keys in payment API requests to ensure duplicate retries don't result in double charges.
FAQ
Are UUIDs truly unique?
Practically yes. The probability of two random v4 UUIDs colliding is 1 in 2^122. You'd need to generate 2.7 billion UUIDs per second for 100 years to have a 50% chance of a single collision. For any realistic application, you can treat them as unique.
UUID vs GUID — what's the difference?
They're the same thing with different names. UUID is the IETF/RFC standard term; GUID is Microsoft's term, commonly used in .NET and Windows contexts. The binary format is identical.
Can I use a UUID as a password?
No. UUIDs are unique but not secret, and their format is predictable (v4 always has "4" in position 13 and one of "8", "9", "a", or "b" in position 17). Use a proper password generator that produces strings with more entropy and less structure.
Does the UUID generator store generated IDs?
No. UUIDs are generated client-side using the Web Crypto API. Nothing is sent to a server. Generated IDs exist only in your browser until you navigate away.
Are the generated UUIDs truly unique?
UUID v4 generates 122 bits of cryptographic randomness. The probability of a collision is astronomically small. In practice, UUID v4 values are considered unique for all real-world applications.
By the Numbers
- UUID v4 contains 122 random bits — collision probability is ~1 in 5.3 × 10^36, effectively zero for any practical use
- Approximately 1 billion UUIDs are generated across the global internet daily (estimated from database usage patterns)
- RFC 9562 (2024) updated the UUID standard to add v6, v7, and v8 for time-ordered and custom use cases
- UUID v7 uses Unix Epoch milliseconds in the most significant bits, making UUIDs sortable by creation time for database indexing