How large a file can I open — is it uploaded?

Nothing is uploaded. Only the part you’re looking at is read (64 KB chunks, with at most 16 MB cached), so multi-gigabyte files open fine. Search and compare scan the whole file in chunks; speed depends on your disk and browser, typically hundreds of MB per second.

No — it’s a read-only viewer and never modifies the original. To extract part of a file, select it and export it as hex, Base64, a C array or another format.

How accurate is file type detection?

Detection uses magic numbers at the start of the file (or at fixed offsets for formats like TAR and ISO), which is very reliable for formats with clear signatures such as PNG, PDF, ZIP and ELF. ZIP-based formats like docx, xlsx, apk and jar are told apart by the file names inside the archive. Weak signatures such as MP3 frames, zlib streams or JSON are listed last as hints.

What’s the difference between little- and big-endian?

Multi-byte numbers can be stored in two byte orders. Little-endian (LE) puts the least significant byte first and is used by x86, ARM, Windows and most file formats; big-endian (BE) puts the most significant byte first and is used by network protocols, Java class files, PNG and JPEG. For example, 01 02 is 513 as LE and 258 as BE. If unsure, check both and see which value makes sense.

🔦

🔦Online Hex Viewer - Inspect, Search & Compare Binary Files

See what’s really inside a file, confirm its true format or find where two files differ — without installing anything. Drop a file in to browse it as classic hex + ASCII with 8, 16 or 32 bytes per row and offsets in hex or decimal. The file is read in small chunks on demand and the view is virtualized, so multi-gigabyte disk images and logs open instantly and scroll smoothly. Jump to any offset, or search for a hex byte sequence or UTF-8 / UTF-16 text (optionally case-insensitive) forwards and backwards. Click or drag to select bytes and the data inspector shows int8–int64, unsigned, float16/32/64, LEB128, Unix timestamps, Windows FILETIME, DOS dates, GUIDs and IPv4 in little- or big-endian, plus the selection as UTF-8, UTF-16 or Windows-1252 text. Magic numbers identify more than 60 file types — images, audio and video, archives, Office files, executables, fonts, databases and more — with a warning when the extension doesn’t match. PNG chunks, JPEG segments, GIF, BMP, ZIP local headers and end of central directory, PDF header and trailer, ELF headers, PE DOS / COFF / optional headers and section tables, and RIFF / WAV chunks are color-coded with their field values listed; click to jump. Compare mode shows two files side by side with differing bytes in red and steps through each difference. Export a selection as hex, hexdump, Base64, a C array, Python bytes or a JS Uint8Array. Strictly read-only, and nothing is uploaded.

Free · No sign-up · Runs in your browserLast updated: 2026-10-26

Loading tool...

How to use Hex Viewer

1Drop a file in or click to choose one (add a second file to compare)
2Scroll, jump to an offset or search bytes / text, and click bytes to inspect values
3Check the detected file type and structure fields, and export selections as hex, Base64, a C array and more

FAQ

How large a file can I open — is it uploaded?: Nothing is uploaded. Only the part you’re looking at is read (64 KB chunks, with at most 16 MB cached), so multi-gigabyte files open fine. Search and compare scan the whole file in chunks; speed depends on your disk and browser, typically hundreds of MB per second.
Can I edit the file?: No — it’s a read-only viewer and never modifies the original. To extract part of a file, select it and export it as hex, Base64, a C array or another format.
How accurate is file type detection?: Detection uses magic numbers at the start of the file (or at fixed offsets for formats like TAR and ISO), which is very reliable for formats with clear signatures such as PNG, PDF, ZIP and ELF. ZIP-based formats like docx, xlsx, apk and jar are told apart by the file names inside the archive. Weak signatures such as MP3 frames, zlib streams or JSON are listed last as hints.
What’s the difference between little- and big-endian?: Multi-byte numbers can be stored in two byte orders. Little-endian (LE) puts the least significant byte first and is used by x86, ARM, Windows and most file formats; big-endian (BE) puts the most significant byte first and is used by network protocols, Java class files, PNG and JPEG. For example, 01 02 is 513 as LE and 258 as BE. If unsure, check both and see which value makes sense.