How do I get the date from a Discord ID?

Discord IDs are Snowflakes. Shift out the low 22 bits to get the millisecond offset, then add Discord's epoch of 1 January 2015. This decoder does both steps when you select the Discord epoch.

Why does my Snowflake decode to the wrong date?

Wrong epoch. Snowflake timestamps are offsets from a service-specific starting point, so a Twitter ID decoded with Discord's epoch is off by more than four years.

What is the difference between ULID and UUID v7?

Both put a millisecond timestamp first and sort by creation time. ULID uses a 26-character Crockford base32 text form; UUID v7 keeps the standard 36-character UUID format and fits existing UUID columns.

ULID / Snowflake ID Decoder

Extract the timestamp, shard and sequence from ULIDs and Snowflake IDs.

live · client-sideIdentification & Serialsfree · no sign-up

RAW INPUT

Snowflake epoch

DECODED

Output appears here as you type — decoded on your device.

Input is processed on your device. Nothing you paste is sent anywhere.

What is the ULID / Snowflake ID Decoder?

A Snowflake/ULID decoder extracts the embedded creation timestamp from a time-sortable identifier — a 64-bit Snowflake ID or a 26-character ULID — along with the worker and sequence bits that make it unique.

Snowflake IDs were designed at Twitter to generate sortable, unique 64-bit identifiers without central coordination. The layout is 41 bits of millisecond timestamp, 10 bits of machine or worker identifier and 12 bits of per-millisecond sequence. The timestamp is an offset from a custom epoch, which is why decoding requires knowing whose Snowflake it is: Twitter's epoch is 4 November 2010, Discord's is 1 January 2015.

ULIDs solve the same problem in a text-friendly way: 26 Crockford base32 characters, of which the first 10 encode a 48-bit millisecond timestamp and the remaining 16 are randomness. They sort lexicographically in creation order, are URL-safe, and are case-insensitive by design.

Both are worth decoding for the same reason: any record keyed this way carries its own creation time, so you can date a message, an order or a log entry from its identifier alone.

ULID / Snowflake ID Decoder at a glance

Snowflake: 64 bits: 41 timestamp · 10 worker · 12 sequence
Epochs: Twitter 2010-11-04 · Discord 2015-01-01
ULID: 26 chars Crockford base32: 48-bit time + 80-bit random
Both: Lexicographically sortable by creation time

How to use the ULID / Snowflake ID Decoder

1Paste the Snowflake ID or ULID.
2For a Snowflake, select the matching epoch — Twitter, Discord or custom.
3Read the recovered creation timestamp, worker ID and sequence.

When you would reach for it

Dating a Discord message or user account from its ID.
Correlating log entries by extracting timestamps from record IDs.
Verifying that your own ID generator encodes the epoch you configured.

ULID / Snowflake ID Decoder — frequently asked questions

Last reviewed 25 July 2026 · questions or corrections: contact@decoder.tools