How to Generate One or More ULIDs

What the ULID Generator does

The ULID Generator creates identifier strings that combine a time-derived portion with a randomly generated portion. It is useful when you need one identifier for a record, file, request, or other item, or when you need a batch of identifiers for a test dataset or setup task.

The tool accepts an optional count so you can request one identifier by default or a larger batch. Each successful identifier has a consistent 26-character format: a 10-character time component followed by a 16-character component encoded with a base32 alphabet. This structure makes the output easy to copy into systems that accept text identifiers.

Step-by-step instructions

Open the ULID Generator and review the available count control. If you need one identifier, leave the optional count unchanged. The default count is 1.
To request multiple identifiers, enter a numeric count in the count control. Values from 1 through 100 are supported by the generation path. A value below 1 is replaced with 1 before generation, while a value above 100 is replaced with 100.
Start the generation operation. The tool creates each identifier by combining a time-derived component measured in milliseconds with a random component. The random component comes from operating-system randomness.
Review the returned identifiers and the applied count. If you requested a batch, use the complete returned collection rather than relying on the displayed sample. The result also includes a timestamp and identifies the first generated identifier as a sample.
Copy the identifiers you need into the destination system. Keep the identifiers as text so their 26-character form and base32 characters remain unchanged.

How to read the result

Use the applied count to confirm how many identifiers the operation used for the result. This matters when the requested count is outside the supported range, because the tool adjusts it before generation. The timestamp describes the returned operation, while the sample is only the first generated identifier, not a replacement for the rest of a batch.

Check the format of an individual result by counting 26 characters and looking for the two-part structure: 10 characters for the time-derived component and 16 characters for the randomly generated component. A successful result contains the identifiers along with the applied count, timestamp, and first-item sample.

If computation encounters an exception, the operation reports failure and includes the exception text. Treat that response as a failed generation rather than using a partial value.

Worked example

A developer needs three text identifiers for three newly created test records.

Enter 3 as the count and start generation. Review the full returned collection, confirm the applied count, and copy the identifiers you need.

The successful result contains three identifier strings, an applied count of 3, a timestamp, and the first of the three identifiers as the sample.

Limitations

The runtime contract declares network access as false, but that declaration does not establish where the tool runs or how generated values are handled after display.
The documented behavior covers the numeric count path. Handling of strings, null values, or other non-numeric count inputs is not established.
The documented generation method does not establish that identifiers are globally unique or ordered across separate operations.

Common errors

Using the first-item sample as though it were the full batch can omit requested identifiers. Use the returned identifier collection and check the applied count before copying the results.

FAQ

Can I generate one ULID at a time?

Yes. Leave the optional count at its default to generate one identifier; the result includes that identifier, the applied count, a timestamp, and the first-item sample.

How many ULIDs can I generate in one operation?

You can request a numeric count from 1 through 100. A value below 1 is replaced with 1, and a value above 100 is replaced with 100 before generation.

What does a generated ULID look like?

Each successful identifier is 26 characters long, with a 10-character time component and a 16-character base32-encoded component made from 10 random bytes.