UUID Generator: What is a UUID and When Should You Use One?

UUIDs appear in every codebase, but developers often use them without fully understanding when they're the right choice — and when they're not. This guide explains what UUIDs are, the different versions, and the practical tradeoffs for using them as database primary keys.

What is a UUID?

A UUID (Universally Unique Identifier) is a 128-bit number, standardised in RFC 4122, represented as 32 hexadecimal digits in 5 groups separated by hyphens:

a3f8b2c1-4e5d-6f7a-8b9c-0d1e2f3a4b5c
^^^^^^^^ ^^^^ ^^^^ ^^^^ ^^^^^^^^^^^^
8 chars  4    4    4    12 chars

The "universally unique" part means two independently generated UUIDs should never collide — the probability of a collision is astronomically small (about 1 in 2^122).

UUID Versions

v4 — The Default Choice

UUID v4 is purely random and the most commonly used version. It's what most libraries generate by default and what Dummy JSON Generator produces for the UUID field type.

// JavaScript (native, no library)
const id = crypto.randomUUID();
// "a3f8b2c1-4e5d-4f7a-8b9c-0d1e2f3a4b5c"
// Note: version digit is always 4 (the first char of the 3rd group)

v7 — Better for Database Primary Keys

UUID v7 encodes a millisecond timestamp in the first 48 bits, making it lexicographically sortable. This dramatically improves B-tree index performance compared to v4:

npm install uuidv7

import { uuidv7 } from 'uuidv7';
const id = uuidv7();
// "018f4a2b-c3d4-7e5f-a6b7-8c9d0e1f2a3b"
// First segment encodes creation time — rows insert in order

UUID vs Auto-Increment ID — Which Should You Use?

The practical recommendation: Use UUID v7 (or ULID — a similar alternative) when you need globally unique IDs. Use auto-increment BIGINT when you have a single database, don't need to merge data from multiple sources, and want maximum query performance.

Generating UUIDs in Different Languages

// JavaScript / TypeScript (native)
crypto.randomUUID()

// Node.js (native)
import { randomUUID } from 'crypto';
randomUUID()

// Python
import uuid
str(uuid.uuid4())

// PHP
Str::uuid()  // Laravel
// or: sprintf('%04x...', ...) for vanilla PHP

// Go
import "github.com/google/uuid"
uuid.New().String()

// Java
UUID.randomUUID().toString()

// C# / .NET
Guid.NewGuid().ToString()

// SQL (PostgreSQL)
gen_random_uuid()  -- or uuid_generate_v4() with pgcrypto

// SQL (MySQL 8+)
UUID()

Storing UUIDs in Databases

PostgreSQL

Use the native uuid type — stores as 16 bytes, not 36-character text:

CREATE TABLE users (
    id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
    email TEXT NOT NULL
);

MySQL

Store as BINARY(16) for efficiency, or CHAR(36) for readability:

CREATE TABLE users (
    id BINARY(16) PRIMARY KEY DEFAULT (UUID_TO_BIN(UUID(), 1)),
    email VARCHAR(255) NOT NULL
);
-- UUID_TO_1 reorders bytes to be time-ordered (like v7)

SQLite

Store as TEXT — SQLite has no native UUID type:

CREATE TABLE users (
    id TEXT PRIMARY KEY,
    email TEXT NOT NULL
);

Generate Bulk UUIDs for Testing

Need hundreds of UUIDs for test data? Dummy JSON Generator generates UUID fields as part of a larger dataset — set the UUID field as your ID column, add other fields, and generate as many records as you need. Each record gets a unique v4 UUID automatically.

For just a list of UUIDs with no other fields, generate a single-field schema with uuid type and export as JSON or CSV.