Canvas Fingerprint Test

Canvas fingerprinting is a tracking technique that asks your browser to draw a carefully-chosen image in an invisible HTML canvas and then reads back the pixel data. Because font rendering, anti-aliasing, emoji glyphs, and GPU compositing vary between operating systems, browsers, and graphics drivers, the resulting pixel buffer is usually unique to a single device. Hashing that buffer yields a stable identifier that survives cookie deletion and incognito mode.

Yes — the render recipe here (Arial text, overlapping circles with multiply blending, Times New Roman fallback) is deliberately similar to the instructions used by commercial fingerprinting libraries like FingerprintJS, so the SHA-256 hash you see is a good proxy for what a tracker would record about your browser. The hash changes if you switch browsers, upgrade your OS, or change your GPU driver.

WebGL exposes your GPU to web pages so they can render 3D graphics, but it also reveals low-level details: the driver vendor, the exact GPU model string, shading language version, and a long list of supported extensions. Bundled together and hashed, these values form a second fingerprint that is often even more unique than the 2D canvas fingerprint — especially on desktops where the GPU varies enormously.

The UNMASKED_VENDOR_WEBGL and UNMASKED_RENDERER_WEBGL parameters return the exact GPU model — for example, 'ANGLE (Apple, Apple M2 Pro, OpenGL 4.1)'. Firefox hides these behind privacy.resistFingerprinting; Safari blocks them by default; Chrome reveals them unless you install a fingerprinting-resistant extension. When they show as 'blocked' here, your browser is actively defending against this fingerprint.

Turn on Firefox’s privacy.resistFingerprinting (about:config) or use Tor Browser — both return a uniform, fake pixel output. Brave randomizes per-session. In Safari, Intelligent Tracking Prevention makes canvas readback noisier and restricts third-party usage. Chrome users can install the CanvasBlocker extension. All of these will change the hash you see on this page on every reload, which is exactly what defeats the tracker.

Not quite every browser, but close. On desktop, canvas + WebGL hashes together are unique for roughly 95–99% of visitors in published studies. Mobile devices cluster a bit more (many iPhones of the same model produce identical hashes) but combined with other fingerprint signals they’re still highly identifying.