Device Fingerprint

See what your browser reveals about your device and the fingerprints websites can compute from it. 100% in your browser.

What is device fingerprinting?

Every browser exposes small details about the device it runs on: screen size, GPU model, installed fonts, time zone, CPU cores, and dozens more. None of them identifies you on its own, but combined they form a fingerprint that is often unique enough to recognize a returning device without cookies. This tool reads the same signals a fingerprinting script would and shows exactly what your browser reveals. Everything on this page is computed locally in your browser: nothing is uploaded or stored.

Fingerprints are used both to track visitors across sites and to fight fraud, since bots and spoofed browsers often leak inconsistencies between these signals. Values can change when your setup changes (a browser update, a new monitor, different fonts), which is why real-world fingerprinting combines many weak signals instead of relying on a single one.

How each hash is computed

Audio

Plays a fixed oscillator through a compressor in an offline audio context and hashes the rendered samples, which vary with each audio stack's floating point processing.

Canvas

Draws text, gradients, and shapes on an invisible canvas, then hashes the resulting pixels with SHA-256. GPUs, drivers, and font rasterizers all round differently, so identical drawing commands produce different pixels on different machines.

Device

A SHA-256 over every signal shown on this page combined with the audio, canvas, and WebGL hashes. Change any single signal and it changes completely.

WebGL

Hashes what your graphics stack declares about itself: the GPU vendor and renderer strings, the reported limits, and the list of supported extensions.

Fingerprints this page cannot see

Some fingerprints are computed from the connection itself, before any page loads. They are visible only to the server terminating the connection, never to JavaScript, so an in-browser tool cannot measure them. This page deliberately shows only what any website can read from a loaded page.

FingerprintBuilt fromWhat it identifies
JA3 / JA4 The TLS handshake: cipher suites, extensions, elliptic curves, ALPN protocols The TLS library, popular for spotting bots that fake a browser user agent
HTTP/2 (Akamai) Connection settings and frame ordering The HTTP client implementation
TCP/IP (p0f) Packet-level behavior: TTL, window sizes, TCP options The operating system's network stack

Try different tools

Explore all