test-board: liverating-lab v1 // updated 2026-08-18
Methodology

CreepJS vs Pixelscan vs BrowserLeaks vs Iphey: what each one checks

6 min readupdated 2026-08-18
TL;DR

What each fingerprint checker actually tests, and when to use which one.

Every antidetect browser roundup name-drops the same four fingerprint checkers, usually without explaining that they test different things. Here's what each one is actually for.

CreepJS

The most thorough public probe. Instead of a single pass/fail, it outputs a "trust score" based on cross-checking dozens of signals against each other — does your reported timezone match your IP-derived location, does your GPU vendor string match what that GPU can actually render, does your hardware concurrency count match your claimed device class. A tool can spoof every individual value and still score badly here if the values contradict each other.

Pixelscan

Narrower scope, more visual. It's built around canvas/WebGL rendering consistency and automation-flag detection (things like navigator.webdriver being left exposed). It's the closest free approximation of what a mid-tier ad-platform anti-fraud check looks like, which is why it's the most commonly screenshotted checker in vendor marketing.

BrowserLeaks

Not a score — a signal-by-signal report. IP/DNS leak detection, WebRTC leak detection, font enumeration, audio context fingerprinting, each broken out separately. Most useful when a profile is failing and you need to find which specific layer is leaking, rather than confirming that everything's fine.

Iphey

A second full-consistency opinion, similar in spirit to CreepJS but with different internal weighting. Useful specifically when CreepJS and Pixelscan disagree on a profile — Iphey acts as a tiebreaker rather than a primary tool.

Which one should you actually trust?

None of them alone. A profile that passes Pixelscan but fails CreepJS's consistency check is a profile with a good-looking canvas hash and a contradictory GPU/hardware story underneath — exactly the kind of thing a sophisticated anti-fraud system (which none of these four tools fully replicate) is built to catch. Run all four, and treat disagreement between them as more informative than agreement.

Practical note: passing every public checker is necessary but not sufficient. These tools approximate what a mid-tier platform checks — they don't replicate proprietary anti-fraud systems at Google, Meta or major exchanges, which layer in behavioral signals no public tool can test.