When you're checking your network environment, most people only look at their current public IP and overlook the network address information the browser itself produces. Especially in web apps that use WebRTC, the browser can pick up several different types of network addresses through ICE candidates.
So a clean result from a plain IP lookup doesn't mean the WebRTC check will line up. If you want to know exactly which addresses your browser is exposing right now, it's still worth running a dedicated WebRTC leak test.

WebRTC (Web Real-Time Communication) is the real-time comms tech built into browsers. Video meetings, in-page voice, live chat support — any of them may be using it.
To set up a real-time connection, WebRTC gathers network candidates through ICE, and those can include local network addresses, public mapped addresses and IPv6 addresses. What a page is actually allowed to read depends on the browser's version, its network configuration and its WebRTC policy.
Which means WebRTC isn't a feature designed to "expose your IP". An address turning up in the results is really just a side effect of the network probing that real-time connections depend on.
Open a test tool and you'll usually see a few categories worth paying attention to.
This is the public address that internet services can see. If WebRTC reports a public IP that doesn't match what you'd expect from your network's exit point, it's worth digging into.
These usually turn up as private ranges like 192.168.x.x and 10.x.x.x. They only exist inside a LAN, so they aren't a public leak in themselves — but they do tell you which network candidates the browser picked up.
This is the one people miss most. Plenty of networks run IPv4 and IPv6 side by side: your everyday IPv4 check looks clean, yet the WebRTC test may still surface an extra IPv6 address. Best to confirm it separately.
Put simply: don't just ask "did my IP change?" Ask which addresses appeared, what type each one is, and whether they match the network you're actually on.
| Method | How to do it | Who it suits | What to watch out for |
|---|---|---|---|
| Online WebRTC test | Open the test page, let the browser finish scanning, then read the Public IP, Local IP and IPv6 results | Easiest route for regular users and ops teams — usually no extra software to install | If the tool also reports DNS and browser fingerprint, look at those too |
| Browser DevTools | Hit F12, then inspect the network candidates through RTCPeerConnection and ICE candidates | Users with some technical background | You see lower-level detail, but it's a steep read — regular users don't need to study the code |
| Browser settings or extensions | Use the WebRTC-related options your browser offers to restrict part of the network info gathering | Users who need to change browser behaviour | Changes can affect web meetings, voice chat and live co-hosting — rerun the WebRTC test after adjusting |
If all you want is a quick answer on which WebRTC network addresses your browser currently has, ToDetect is one of the more convenient online tools for the job.
Its strength isn't handing you a simple "normal" or "abnormal" verdict — it lays out the detected Public IP, Local IP and IPv6 directly. So when different addresses come up, you can work out whether each one is public, internal or IPv6, instead of staring at a final score.
ToDetect also reports DNS leaks, browser fingerprint and other related data. For anyone troubleshooting a network environment, reading those together surfaces problems far more reliably than a single public IP lookup.
That said, no online tool replaces an actual check of your browser and network configuration. When results look off, read them against the network you're on, your browser version and your IPv4/IPv6 status.
Not always. Say a plain web page reports one public IP and the WebRTC test reports another — start by confirming which network path and address type each one corresponds to. Some of that gap comes from network configuration, parallel IPv4/IPv6 connections, or how the browser itself handles things.
What genuinely deserves attention is whether the results contain a public address you didn't want exposed, or an IPv6 address that clearly doesn't fit your current network.
Seeing only private ranges like 192.168.x.x or 10.x.x.x doesn't automatically mean a public IP leak either. So judging a WebRTC leak isn't about counting IPs — it's about address type and real network context.
First, test on the network you actually use. Switch Wi-Fi, change networks or enable IPv6 and the results can all shift.
Second, don't test IPv4 only. Plenty of devices now run IPv4 and IPv6 together, so confirm both.
Last, always retest after changing browser settings. A config change doesn't guarantee the outcome you wanted — the actual test result is what counts.
A WebRTC leak test isn't complicated. The key is not to fixate on a single public IP. Read Public IP, Local IP and IPv6 separately, then factor in DNS and the browser environment — that usually pins down the problem faster.
For routine checks, one full WebRTC run on ToDetect is enough. The point isn't chasing some fixed result, it's confirming that the addresses shown actually match your real environment.