How to test your Brave browser fingerprint

Updated October 17, 2026 · Independent guidance

Open the test in Brave to inspect the signals available to this page. Brave documents fingerprint randomization as part of Shields. A readable or stable value does not by itself mean that protection has failed.

A repeatable Brave test

1. Record your Brave version, operating system and this site's Shields settings. Menu labels and available protections can vary.
2. Open the browser fingerprint test in Brave and wait for the raw output.
3. Record the canvas digest, WebGL information, languages and estimated font results.
4. Rescan with the same settings to compare observations within the session.
5. If you want to compare sessions, close and reopen the browser, return to the same URL and record the same rows again.

Keep the device and display setup fixed during comparison. This test does not save previous runs, so retain your own notes if you need a history. Opening a new tab alone should not be treated as a controlled restart experiment.

Why can Brave return different values?

Brave's Shields documentation describes modifying or restricting some fingerprintable APIs. Its documented randomization behavior can vary values by site and browser session. That makes a changed canvas digest compatible with a privacy defense, but this tool cannot prove which mechanism caused a particular change.

Why did my canvas digest stay the same?

A stable result over two runs only establishes stability for the tested image under those conditions. A same-session rescan and a new-session comparison ask different questions. Our single-origin test cannot measure whether an unrelated website receives the same output.

Inspect the full signal set rather than treating one row as a pass or fail. The WebGL rows read vendor and renderer strings; they do not exercise every graphics API. The font estimate probes a fixed list and can be influenced by fallback rendering.

Why might the browser label look like Chrome?

This site's browser label is parsed from the reported user-agent string. It is not an independent identification of the installed application. If the string presents Chrome-compatible information, the displayed name can reflect that. Use Brave's own version information to label your experiment.

What this check cannot certify

It does not measure all of Shields, audit blocked trackers, calculate anonymity or rank Brave against other browsers. The public IP panel is a separate network observation, and optional WebRTC testing contacts an external STUN service only when you request it. Read the data-flow disclosure before using that optional check.

Use the results interpretation guide to understand warnings. For a different protection strategy, read the Tor Browser guide. This is an independent tool and is not affiliated with Brave.