UUID v7 Generator (Time-Ordered IDs)

UUID v7 generator online free (time-ordered, RFC 9562)

For twenty years, the humble UUID v4 was the default answer to “I need a unique ID”: 128 bits of pure randomness, easy to generate, impossible to collide. But randomness has a cost. When v4 IDs become database primary keys, every insert lands at a random position in the B-tree index, forcing page splits, fragmentation, and slower writes. UUID v7 — the newest version, standardized in RFC 9562 in May 2024 — fixes this by embedding the current Unix timestamp in the first 48 bits. The IDs stay globally unique, but they also sort in creation order, so inserts append instead of scattering.

This free generator produces both versions in bulk: up to 1,000 IDs per batch, with hyphenated or compact formatting and optional uppercase output. Everything runs on the Web Crypto API inside your browser — no server, no signup, no tracking of your IDs. The built-in inspector decodes any UUID you paste: version, variant, and — for v7 — the exact date and time baked into the ID.

v4 vs v7: a quick comparison

Why time-ordered matters for databases

Database indexes are usually B-trees, which are happiest when new keys arrive in order: each insert lands at the rightmost leaf and the tree grows cleanly. Random v4 keys do the opposite — every insert hits a random leaf, splits pages, and leaves half-empty pages behind. On write-heavy tables this measurably slows inserts and bloats the index. UUID v7 behaves like a well-designed sequential key while keeping the distributed-generation superpower of UUIDs: any server, anywhere, can mint IDs without coordinating with a central counter.

The ecosystem has noticed. PostgreSQL 18 added native UUID v7 generation, Python 3.14 ships uuid.uuid7(), and major frameworks are adopting v7 as the default. If you are starting a new project and reaching for UUID primary keys, v7 is the modern answer — keep v4 for the cases where you explicitly want no time information in the ID.

How bulk generation works here

Choose v7 or v4, pick a preset (1, 10, 50, 100) or type a custom count up to 1,000, and hit Generate. Each ID is rendered as a clickable row — click any single UUID to copy it, or use Copy all to grab the whole batch. The Download .txt button exports the list as a plain text file, handy for seed scripts and fixtures. The format controls apply live: switch between hyphenated and compact (no dashes) or toggle uppercase without regenerating.

Bulk generation is a daily need in real workflows: seeding a staging database with ten thousand rows, generating API keys for a load test, or creating fixture IDs for integration tests. Because every ID is minted with the Web Crypto API, the output is suitable for real systems — not just placeholder text. The compact format (32 hex characters, no dashes) is popular in URL slugs and log correlation IDs where every character counts.

Decoding a UUID with the inspector

Paste any UUID into the inspector to read what is inside it. The version nibble (the 13th hex digit) reveals the UUID version — 4 for random, 7 for time-ordered, 1 for the old timestamp+MAC scheme. The variant bits tell you which standard family it belongs to (almost always RFC 9562). And if it is a v7 ID, the inspector decodes the 48-bit embedded timestamp into the exact date and time the ID was created — a quick way to audit when records were minted.

Privacy note

All generation and inspection happen locally in your browser via the Web Crypto API — nothing is sent to a server. One thing to remember: v7 IDs visibly encode their creation time, so anyone who sees the ID can tell when it was made. If that matters for your use case (public-facing tokens, for example), v4's pure randomness is the safer choice.

Related developer tools

Pair this with ToolsOfWeb's Regex Tester to validate ID formats in your codebase, or the JSON Formatter when your API payloads carry these IDs around.

FAQs