Snowflake ID Decoder

A snowflake is a 64-bit ID used by Discord, Twitter/X and Instagram. The top bits hold a millisecond timestamp offset from a fixed platform epoch, and the lower bits hold a worker id, process id and a per-millisecond sequence counter. That makes IDs sortable by creation time without a separate timestamp column.

The timestamp stored in a snowflake is counted from a service-specific epoch, not the Unix epoch. Discord starts at 2015-01-01, Twitter at 2010-11-04 and Instagram at 2011-08-24. Choosing the right platform applies the right offset so the decoded time is correct. Use a custom epoch for your own snowflake implementation.

No. Decoding happens entirely in your browser using BigInt math. The ID never leaves your device, so it is safe for IDs that reference private accounts, channels or messages.

The timestamp is precise to the millisecond and reflects when the ID was generated, which is effectively when the account, message or post was created. It is read directly from the ID bits, so there is no rounding.

They identify which machine generated the ID and the order of IDs created within the same millisecond. Worker (or datacenter) id and process (or machine) id are each 5 bits (0-31), and the sequence is 12 bits (0-4095). They are mostly useful for debugging ID generation, not for end users.

A snowflake is up to 64 bits, but JavaScript numbers lose precision above 2^53. Parsing the ID as a BigInt keeps every bit intact so the timestamp and the worker, process and sequence fields are extracted exactly.