ULID Generator

How to use

1. 1Choose how many ULIDs you want to generate (1 to 100).
2. 2Pick a timestamp source: Now uses the current time, Unix ms takes a specific integer, and Pick a date converts a local date and time.
3. 3Leave Monotonic mode on (recommended) so values sharing a millisecond stay strictly increasing, or switch it off to draw fresh randomness for every value.
4. 4Click Generate ULIDs to produce a fresh batch. Use the per-row Copy button to copy one identifier, or Copy all to copy the whole list.
5. 5Paste any ULID into the Decode any ULID field to see its embedded creation time and the 80-bit random portion.

About this tool

ULID Generator creates Universally Unique Lexicographically Sortable Identifiers in your browser, following the public ULID specification. Each value is 128 bits long, exactly like a UUID, but laid out so the first 48 bits hold a Unix-epoch millisecond timestamp and the last 80 bits hold cryptographically random data from crypto.getRandomValues, the same CSPRNG that backs crypto.randomUUID and SubtleCrypto in modern browsers. ULIDs are serialized as 26 characters of Crockford Base32, which deliberately omits the letters I, L, O, and U so a printed value never gets misread as 1, 0, or anything offensive, and which preserves lexicographic order so plain string sorting also sorts by time. Pick a batch size up to 100 and generate using the current time, a specific Unix millisecond value, or a date and time you choose, so every identifier in the batch shares the same 10-character prefix and is guaranteed to sort together. Monotonic mode follows the ULID spec recommendation: when several values share the same millisecond, the random component is incremented by 1 instead of being regenerated, which keeps the IDs strictly increasing inside a single millisecond and matches how the original Stripe ULID library, ulid-py, ulidx, and oklog/ulid all behave. A lowercase toggle switches the output if your downstream tools prefer lowercase. The built-in decoder accepts any 26-character ULID, tolerates whitespace and stray hyphens from copy-paste, applies the Crockford alphabet aliases for I/L (treated as 1) and O (treated as 0), and reports the embedded creation time as a Unix-ms integer and an ISO 8601 string, plus the 80-bit random portion in hexadecimal. Useful as application primary keys, event IDs, trace IDs, log correlation IDs, S3 object keys, idempotency tokens, ledger entry IDs, and any other situation where you want a short, URL-safe, time-sortable identifier without the 36-character footprint of a hyphenated UUID. ULID predates UUID v7 (RFC 9562, May 2024) but solves the same database-friendly-primary-key problem, with a shorter wire format and slightly more random bits per value (80 vs 74). Generation, decoding, and all formatting run entirely in your browser. The identifiers and any ULID you paste into the decoder never leave your device.

Free to use. Works in your browser. No signup, no login.