Explainer · the fundamentals

What is my IP address?

Your IP is the single most important identifier your device has on the internet — and the only one the rest of the network uses to reach you. Here is what it actually means, what websites learn from it, and how to change it.

See my IP now →Networking glossary

An IP address is the return label on every packet you send

Whenever your computer talks to a server — loading a web page, fetching email, sending a message — it wraps every chunk of data into a small envelope called a packet. The packet carries the destination it is heading to and the address it should be replied to. That reply-to address is your IP address. Without it, the server would have no way to send the response back. Everything else on the internet — DNS names, TLS certificates, account logins — is layered on top of this fundamental routing fact.

IPFerret shows you the IP the rest of the internet actually sees, on the home page, in under a second. No sign-up, no profile, and you can verify it matches the API output at /api/ip.

Public IP vs. private IP — they're both yours

Almost every device today actually has two IP addresses: one for the local network it sits on, and one for the wider internet.

The private IP is assigned by your router and is meaningful only inside your home or office. It looks like one of:

These ranges are reserved by RFC1918 and are not routable on the public internet. Every home in your neighborhood likely has a device at 192.168.1.1; they don't conflict because that address is meaningful only inside each home's own router.

The public IP is what your ISP assigns to your router. There is exactly one of these per household (usually), and it is the address every site sees when traffic arrives. Your router does the translation between the two using a process called Network Address Translation (NAT): outbound traffic gets rewritten to use the public IP plus a unique port number; inbound replies arrive at the public IP and the router knows which internal device to forward them to.

IPv4 and IPv6 — why you might see both

The original IP address format, IPv4, uses 32 bits — enough for about 4.3 billion addresses. In 1981 that seemed huge. By the late 2000s, the major regional registries began running out, and today every IPv4 address is either in use or being recycled.

IPv6, designed in the 1990s and slowly rolled out since 2012, uses 128 bits. That gives enough addresses for roughly 7.9 × 10²⁸ per person alive. IPv4 addresses look like 203.0.113.42; IPv6 addresses look like 2606:4700:4700::1111 — four-hex groups separated by colons, with the shorthand :: standing in for one consecutive run of zero groups.

Most modern connections are dual stack: you have one of each. Your device prefers v6 when both ends speak it, and falls back to v4 otherwise. IPFerret shows whichever your browser used to reach us. If you want to verify you have a working v6 path, run the IPv6 test.

What can a website learn from your IP, exactly?

More than nothing, less than people often fear. From an IP address alone, a service can look up:

What the IP does not reveal, just by itself:

These extra pieces only attach to an IP when a service correlates them — for instance, your ISP knowing which customer was leasing the address at a given timestamp, or a site you signed into associating the IP with your account.

Static, dynamic, and CGNAT — which kind do you have?

Most home broadband uses a dynamic IP from a DHCP pool. The lease lasts hours to weeks; when it expires (or your modem reboots), the ISP may hand back a different address. Some ISPs are sticky and keep you on the same one for months at a time, but there is no contract guaranteeing it.

A static IP is reserved for you by the ISP and never changes. Businesses pay extra for this when they need to host services or whitelist their IP somewhere else. Residential static IPs are rare and usually require asking the ISP and paying a monthly fee.

CGNAT (carrier-grade NAT) is the IPv4 conservation trick most mobile carriers and many regional ISPs now use: instead of giving every customer a unique public IPv4, the ISP gives you an address in the 100.64.0.0/10 range and shares one real public IPv4 between many customers via a giant NAT box upstream. From the customer's side, your machine sees a 100.x.x.x address; from a website's side, your traffic appears to come from the carrier's pooled public IP, mixed in with potentially thousands of other subscribers. Read the full CGNAT explainer for the implications.

How to change your IP

Depending on what you're trying to accomplish, you have a few options. From the easiest and least-private to the strictest:

  1. Reboot your modem. Unplug it for five minutes. When it reconnects, your ISP often hands out a different DHCP lease. Works on most home connections but does not change your geographic region or ASN.
  2. Use a VPN. A VPN tunnels your traffic through a server elsewhere; the outside world sees the VPN's exit IP, not yours. Pick a reputable provider and verify it does not leak via the DNS leak test and the WebRTC leak test.
  3. Use Tor. The Tor Browser bundle routes traffic through three relays so the destination site cannot tell what IP originated the request. Slower but much stronger as a privacy guarantee.
  4. Use a different network entirely. Tethering to your phone, walking to a coffee shop, plugging into a friend's connection — each gives you a different IP with different geo, different ASN, and different reputation history.

The trade-off scales with effort: more isolation costs more speed, more configuration, and more attention to leaks. For most use-cases — bypassing a soft geo-block, watching a show that's not in your country, separating two accounts on a service that hates duplicates — a normal reputable VPN is fine. For high-stakes investigative or journalistic work, Tor or a research-grade VPN with rigorous leak testing is the right tool.

Using IPFerret as your IP reference

On the home page you'll see your public IP, the ISP/ASN, the approximate location, whether IPFerret thinks the address belongs to a cloud or anonymizer range, and a fingerprint of the headers your browser sent.

From a script, hit one of the public JSON endpoints:

All three honor CORS for browser fetches, and none log identifying request data — the full policy is on the privacy page.

Frequently asked questions

Why does Google show a different IP from IPFerret?

Either Google is showing you the IP via a CDN node that took a different egress path (Google sees connections from your ISP's edge cache, IPFerret sees the original), or one of the two is reporting your IPv6 while the other is reporting your IPv4. Run /ipv6-test to find out which.

Can my employer see my IP when I work from home?

When you connect to a corporate VPN, the corporate side sees the VPN exit's address, which is the public IP you have at home. They cannot see your LAN IP or what other devices are on your home network. They can usually correlate the home IP with the ISP and a city-level location.

Why does my phone's IP change when I switch from Wi-Fi to data?

Each network has its own IP. On Wi-Fi you get an address from your router (and your home ISP). On cellular data you get an address from your carrier — usually a CGNAT address, with the carrier's pooled public IP visible to external services. The two are different networks; expect different IPs.

Is my IP address considered personal data under GDPR?

The EU's general view is yes — IP addresses can be personal data when they can be linked to an identifiable individual, which on a home connection they often can. That is why IPFerret does not log IPs by default, and why our privacy policy covers the topic explicitly.

Keep exploring