You can't make a browser fingerprint disappear. Every page you load can read some details about your browser and device, and the combination is often distinctive. What you can do is make that fingerprint less useful to whoever collects it, and there are two honest ways to do that: blend in, so your browser looks like a large crowd of identical browsers, or keep identities separate, so each activity runs in its own profile with its own stable values and nothing links them.
In short: You can't remove a fingerprint, only make it less useful. Either blend in (Tor Browser, Firefox's and Safari's protections, Brave's randomization) or keep identities apart in separate profiles with stable, consistent values. Random values on every visit, and piles of single-purpose extensions, often make a browser stand out more.
This guide explains both strategies, what Tor Browser, Firefox, Brave and Safari actually do (each checked on the vendor's own pages on 25 October 2026), why random values on every visit can backfire, and how to test your own setup. If you're new to the topic, start with what browser fingerprinting is and how it works.
Can you stop browser fingerprinting completely?
No. Some details reach every website whether you like it or not, and the rest can be measured by any script the page runs.
The W3C's Mitigating Browser Fingerprinting in Web Specifications separates passive fingerprinting, from what the browser sends in every request (IP address, user agent, language headers), from active fingerprinting, where JavaScript measures more: canvas drawing, WebGL, audio processing, screen size, time zone. You can't stop sending requests, and turning JavaScript off breaks most of the web and is itself an unusual setting.
So the useful question isn't "how do I hide?" but "how do I make my fingerprint either common or compartmentalised?"
What are the two strategies that actually work?
Blending in and separating identities. Everything else is a variation on one of them.
Blending in means reducing how much your browser differs from others. The EFF, which runs the Cover Your Tracks test, describes browsers that protect users by "making all instances of their browser look the same", adding that "there is strength in numbers" (EFF, November 2020). The same post says tools that "randomize the results of fingerprinting metrics" have "the potential to confuse trackers". Tor Browser, Firefox's protections and Safari's lean toward blending in; Brave adds randomization on top.
Separating identities accepts that each browser has a fingerprint and makes sure one fingerprint is used for one purpose only. Your banking and your hobby forum run in different profiles with different storage, connections and device values, so a tracker that sees one learns nothing about the other.
The two can be combined. What doesn't work is a pile of extensions that each change one value at random.
How do privacy browsers handle fingerprinting?
Here is what each vendor says its browser does, checked on 25 October 2026.
| Browser | Approach | On by default? | Main trade-off |
|---|---|---|---|
| Tor Browser | Standardise values so users look alike; letterboxing | Yes | Runs over the Tor network; new add-ons discouraged |
| Firefox | Block known fingerprinters, limit what suspected ones can read | Known fingerprinters in normal browsing; more in Private Browsing and ETP Strict | Some sites may break in Strict |
| Firefox resistFingerprinting | Report standard values (UTC time zone, fixed window sizes) | No | Mozilla doesn't recommend it for most users |
| Brave | Randomise canvas, audio and other outputs per site and per session | Yes | Values differ per site; fingerprint tests may report added noise |
| Safari | Block known fingerprinting companies, add noise, limit screen data | In Private Browsing | Some site features may be affected |
Tor Browser: make everyone look the same
Tor Browser is the clearest example of blending in. Its support page on fingerprinting protections says the goal is "to reduce the number of distinguishable 'buckets' for each metric". The window starts at a size rounded to a multiple of 200 × 100 pixels, and letterboxing adds margins when you resize, so your screen size doesn't single you out. Tor Browser also doesn't let you choose which operating system you appear to be, because "any option to choose would only make users more unique".
The same logic applies to add-ons. The Tor Project says it is "strongly discouraged" to install new add-ons because they "potentially make your Tor Browser fingerprint unique". The crowd only protects you if you stay in it.
Firefox: Fingerprinting Protection and resistFingerprinting
Firefox has two separate features, and they're easy to mix up.
Fingerprinting Protection is part of Enhanced Tracking Protection. Mozilla's support article says Known Fingerprinters Protection is on in normal browsing, and both Known and Suspected Fingerprinters Protection are on in Private Browsing and when Enhanced Tracking Protection is set to Strict. To turn on the suspected-fingerprinter protection everywhere:
- Open Settings, then Privacy & Security.
- Under Enhanced Tracking Protection, choose Custom.
- Tick Known fingerprinters and Suspected fingerprinters, and in the dropdown choose all windows instead of private windows only.
In Firefox 145 (Mozilla blog, 10 November 2025), Mozilla extended these protections to more signals, including fonts, the number of CPU cores, touch points and the size of your dock or taskbar. Mozilla says the protections "cut the percentage of users seen as unique by almost half", and that the new defenses debuted in Private Browsing and ETP Strict first.
resistFingerprinting is the older, stricter mode behind the privacy.resistFingerprinting preference, much of it inherited from Tor Browser. Mozilla's Resist Fingerprinting article lists side effects: the time zone is reported as UTC, new windows open at fixed sizes, and some websites don't work properly, "which is why we do not recommend this feature for the average Firefox user". Mozilla points most users to Fingerprinting Protection instead.
Brave: farbling, or a different fingerprint for each site
Brave takes the randomizing route. Brave's Fingerprinting defenses 2.0 post (May 2020) defines farbling as "slightly randomizing the output of semi-identifying browser features", starting with canvas and Web Audio and extending to WebGL and other APIs. The important detail is the seed: Brave's fingerprinting wiki says the random values come from "a seed that changes per session, per site (eTLD+1) and per storage area".
So a site gets the same value every time it checks during one session, and a different site gets a different value. That stops two sites from matching your fingerprint, without changing the answer every time one site asks. Brave also says it blocks or modifies some APIs "to make Brave instances look as similar as possible", so it combines both strategies.
Safari: Advanced Tracking and Fingerprinting Protection
Apple's Safari User Guide says advanced tracking and fingerprinting protection is on by default in Private Browsing. It blocks connections to data collection companies that use advanced fingerprinting techniques and removes known tracking parameters from links. To use it in all windows, choose Safari → Settings → Advanced and set "Use advanced tracking and fingerprinting protection" to all browsing (Change Advanced settings in Safari). If a site misbehaves, Apple suggests View → Reload Reducing Privacy Protections.
WebKit's Private Browsing 2.0 post (July 2024) explains more of the mechanism: noise is added to canvas pixels drawn by scripts and to audio samples read back by pages, and the reported screen size is fixed to the window's inner width and height.
Why can random values on every visit make you stand out?
Because real browsers are consistent, and noise that changes on every read is a fingerprint of its own.
An unmodified browser draws the same canvas image every time you ask. Its time zone fits its IP address, its user agent matches its client hints, and its window fits inside its screen. When a script reads the canvas twice and gets two answers, or sees a Windows user agent next to a Mac graphics card, the browser has told the site something unusual about itself.
Researchers have measured this. The FP-Scanner study (USENIX Security 2018) detected the fingerprint countermeasures its authors tested by looking for the inconsistencies they introduced: mismatched operating systems across attributes, and tell-tale patterns of added canvas noise. The authors noted that being detected with a countermeasure can make a browser more trackable, though not always.
That's why the serious tools keep values stable. Brave's farbling stays fixed per site for the whole session. Tor Browser gives everyone the same rounded window sizes instead of random ones. The pattern to avoid is the one many extensions produce: one value changed at random on every page load, while everything else stays real.
Why does consistency matter more than rarity?
A rare but consistent fingerprint looks like one real person with an unusual computer. An inconsistent one looks like something is being hidden.
Check these pairs on any browser you care about:
- Time zone and IP location. A VPN that exits in another country while your clock stays at home is the most common mismatch.
- Languages and IP country. At least one browser language should be plausible for the location.
- User agent and client hints. An extension that changes only the user agent string can leave the client hints telling a different story.
- WebRTC and your proxy. If WebRTC reveals a different public IP than the one the site sees, that's a leak.
- Canvas between two reads. It should be the same both times.
The glossary has short definitions of each of these terms if any are unfamiliar.
How do you keep identities separate?
Give each identity its own browser profile and keep each profile's values the same every time it opens.
For personal privacy, the simplest version is two browsers or two profiles: one for accounts you sign in to, another for everything else. That splits cookies and storage, which stops the most common tracking. It doesn't split the fingerprint, though. Profiles in the same browser share the IP address, screen, fonts, graphics card and time zone, as we explain in multiple Chrome profiles vs an antidetect browser.
For work with many accounts, such as an agency managing client accounts, a browser profile manager separates all three layers: storage, network and fingerprint. In Multime, each profile has a seed that drives its masked values (canvas, WebGL image, audio and element sizes), so the profile looks the same every time it opens and two profiles never share values. Time zone, language and location follow the profile's proxy. The limits are worth knowing: screen size is always the real screen, and custom font lists, media devices and speech voices aren't masked. Separation doesn't change what a website's terms allow, so platform rules still apply.
How do you test your own fingerprint?
Run a few targeted tests, change one thing, and run them again.
- Start with consistency. Our browser fingerprint test compares your time zone, languages, formatting locale, WebRTC address, user agent platform and automation flag with the IP address the site sees, and lists any mismatch.
- Check canvas stability. The canvas fingerprint test draws the same image twice, hashes it and tells you whether the result is stable, whether noise was added and whether text measuring still works.
- Check WebRTC behind a proxy or VPN. The WebRTC leak test shows the public address WebRTC reveals and compares it with your connection IP.
- Compare with a crowd. EFF's Cover Your Tracks estimates how your fingerprint compares with recent visitors to its site.
- Read other test sites calmly. Sites such as Pixelscan, BrowserLeaks and CreepJS each check different signals and score them differently. Our guide to fingerprint test sites explains what each one looks at.
Run the tests in a normal window and in a private window, and again after you change a setting, so you can see what the setting actually did.
What doesn't help much?
A few common tips change less than people expect:
- Incognito or private mode alone. It clears cookies at the end of the session but keeps the same device details. Safari and Firefox do turn on extra fingerprinting protection in their private windows, which helps; in other browsers, check what private mode actually changes before relying on it.
- A VPN alone. It changes the IP address and nothing else, and it can create a time zone mismatch.
- Many privacy extensions. Each one changes the browser's behaviour a little, and the Tor Project warns that add-ons can make a fingerprint unique.
- User agent switchers. Changing the string without changing anything else creates exactly the inconsistency described above.