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

How we score fingerprint consistency

Updated quarterlyupdated 2026-08-18
TL;DR

How CreepJS, Pixelscan, BrowserLeaks and Iphey are used to score antidetect browser fingerprint consistency.

Most antidetect browser "reviews" are feature tables copied from each vendor's own pricing page. None of them tell you whether the fingerprint a tool generates actually holds together under scrutiny. This page documents the four tools we run every product through, and what a pass or fail on each one actually means.

Why fingerprint consistency matters more than feature count

An antidetect browser doesn't need to hide who you are — it needs to make a profile look like one coherent device. A fraud system doesn't need to unmask you; it just needs to catch one contradiction between your reported GPU, your canvas hash, and your font list. That's the failure mode we're testing for.

The four tools

CreepJS

The most thorough public fingerprinting probe available. It scores "trust" based on internal consistency between dozens of signals — timezone vs IP-derived location, hardware concurrency vs reported device class, WebGL renderer string vs GPU vendor. A high CreepJS trust score means the spoofed values agree with each other; a low one means they contradict.

Pixelscan

Focused on canvas and WebGL rendering consistency, and on whether the browser's automation flags (like navigator.webdriver) are properly masked. Pixelscan is the closest public approximation of what a mid-tier anti-fraud vendor checks.

BrowserLeaks

A signal-by-signal breakdown rather than a single score — IP/DNS leak detection, WebRTC leak detection, font enumeration, audio context fingerprinting. Useful for finding exactly which spoofed layer is weak, not just that one is.

Iphey

Runs a broad consistency check similar to CreepJS but with different weighting, useful as a second opinion when the two disagree.

Test conditions we hold constant

What we don't do

We don't publish a single "winner." A tool that's excellent for a five-profile agency workflow can be the wrong pick for a 200-profile automation setup, and the fingerprint-consistency test alone doesn't capture that. The scores on our test board feed into the "best for X" comparisons published on the wider network — they're an input, not a verdict.

Status: the live test board on the homepage fills in as each product is run through this pipeline, with the date and product version attached to every score.