ByteScope

File Format Reference

Every format the identifier knows — with its magic number, MIME type and extensions.

Your files never leave your browser — all processing is local.

← Back to the File Type Identifier

Images

Video

Audio

Archives & compression

Documents

Data & databases

Executables & bytecode

Fonts

Text & markup

About this tool

This is the reference side of the File Type Identifier: one page per format, grouped by category. Each format page shows the exact byte signature (the “magic number”) rendered in hex, the canonical MIME type, the file extensions that belong to it, and a short explanation of what the format is actually for.

It's built for the moment you have a mystery file and a hex dump: match the leading bytes against a signature here, and you know what you're looking at. Container formats such as docx, apk and jar are documented too — they're all ZIP files inside, so their page explains the inner directory entries that tell them apart rather than a single header signature.

To identify an actual file rather than look one up, use the identifier itself — drop the file in and it reads the magic number for you, entirely in your browser.

Frequently asked questions

What is a magic number?
It's a short, fixed sequence of bytes at a known position — almost always the very start — that a format writes so tools can recognize it regardless of the filename. PNG uses 89 50 4E 47; a ZIP begins with 50 4B. This reference lists that signature for every format it knows.
Why do docx, apk and jar not have their own header bytes?
Because they are ZIP archives. Their page shows the ZIP local-file header (50 4B 03 04) and then explains the telltale entries inside — a `word/` folder for docx, `AndroidManifest.xml` for apk — that identify the specific format.
How do I identify a specific file?
Open the File Type Identifier and drop the file in. It reads the first bytes locally and matches them against this same database, then links back here for the details. Nothing is uploaded.