IPv4 and IPv6 are the two versions of the Internet Protocol that address
devices on the internet. IPv4 writes four numbers separated by dots, like
203.0.113.45, and its 4.3 billion addresses ran out for new
allocation in 2011. IPv6 writes hexadecimal groups separated by colons, like
2001:db8::1, and is effectively unlimited. Most connections now
carry both.
Run an IP check and you may see two addresses that look nothing alike: one short and dotted, one long with colons. Neither is a mistake, both belong to you, and there is nothing to choose between them.
This guide covers what separates the two versions, why a second one exists, how to read an IPv6 address without guessing, and the few situations where the difference actually decides something.
What's Actually Different Between IPv4 and IPv6?
IPv4 is the addressing system the internet was built on. It uses 32 bits,
written as four numbers separated by dots - 203.0.113.45 - which
works out to about 4.3 billion possible addresses. That was an enormous number
when it was designed, decades before smartphones, smart TVs and everything else
that now expects a connection.
IPv6 uses 128 bits, written as hexadecimal groups separated by colons -
2001:db8::1. Four times the length is not four times the space but
something closer to unimaginable: a single home allocation under IPv6 holds more
addresses than the entire IPv4 internet. Scarcity stops being a design problem, so
the workarounds built around it stop being necessary.
Why Does IPv6 Exist? Wasn't IPv4 Enough?
It genuinely wasn't. The global free pool of unallocated IPv4 addresses was declared exhausted in February 2011, when IANA handed its last blocks to the five regional registries (source: Number Resource Organization). Addresses already in use kept working, and still do - what ended was the supply of fresh ones, leaving only recycled and traded blocks.
That shortage is the reason NAT became standard on every home router: one public address shared by a whole household, which IP address explained: public, private and why it matters covers in full. IPv6 removes the need for it, because there are enough addresses for every device to hold its own.
Most modern connections are dual-stack: they hold both address types and use whichever the other end supports.
How Do You Read an IPv6 Address?
Two shortcuts explain why it looks unreadable. An IPv6 address is eight groups
of four hexadecimal digits joined by colons; leading zeros inside a group can be
dropped, and one run of all-zero groups can be replaced with a double colon. So
2001:0db8:0000:0000:0000:0000:0000:0001 and 2001:db8::1
are the same address written two ways.
What the halves do matters more than the length. The front part is the prefix
and identifies the network: providers typically delegate a home connection
something between a /48 and a /64, and a single
/64 already holds more addresses than the whole of IPv4. The back part
identifies the device on that network, which is the half that can change on its
own.
How Can You Tell Which One You're Using?
Read the format, nothing more. Four numbers with dots is IPv4, groups with colons is IPv6. Open MyIPShow and it shows what your connection presented on arrival, and where a connection carries both, both appear - which is the normal case on most home and mobile networks now.
Whether you have IPv6 at all is your provider's decision, not a setting on your side. Some have rolled it out to every subscriber, some are partway through, and some have not started, so an IPv4-only reading means nothing is wrong with your equipment. How common each version is at any moment is a moving number rather than a fact worth memorising: Google publishes a running measurement of how many of its users arrive over IPv6 (source: Google IPv6 statistics), which is a better answer than any article's snapshot, this one included.
The IP Lookup tool does the same reading for any address you paste in, which is how you check a server or a line other than your own.
How Do IPv4 and IPv6 Compare, Side by Side?
| Property | IPv4 | IPv6 |
|---|---|---|
| Format | 203.0.113.45 |
2001:db8::1 |
| Address length | 32 bits, about 4.3 billion addresses | 128 bits, effectively unlimited |
| Needs NAT to share one address | Yes, on nearly every home line | No, not by design |
| Reachable from outside | Often not, where the provider uses carrier-grade NAT | Each device can hold a routable address |
| Where it stands today | Universal | Widely deployed, not yet everywhere |
Does It Actually Matter Which One You Have?
For browsing, streaming and messaging, no. Your device negotiates the version by itself and you would never know which one carried a given page. It starts to matter in three places:
- Self-hosting and game servers - routers keep IPv6 firewall rules separate from IPv4 port forwarding, so a service that is open on one version can be closed on the other, and each has to be configured on its own.
- Two checkers that disagree - usually one resolved over IPv6 and the other over IPv4, so both are right and they are reporting two different addresses for the same connection. Rule that out before anything else.
- Older IPv4-only services - increasingly rare, and on a dual-stack connection the fallback happens without you noticing.
Which version carries any given request is not your choice either. Browsers run what the standard calls Happy Eyeballs (RFC 8305): they start both connections at once, keep whichever answers first and abandon the other, which is why a broken IPv6 route normally costs a fraction of a second instead of a failed page.
There is one case where the version genuinely decides something. If your provider puts your line behind carrier-grade NAT, nothing from outside can reach you over IPv4 at all, whatever you configure, and IPv6 is the way out because each device can hold its own routable address. Static vs dynamic IP addresses covers how to check whether your line is behind it.
What Does IPv6 Change About Privacy?
Less than the length of the address suggests, and in one respect more. The prefix identifies your connection and stays as long as your provider leaves it alone, so an IPv6 address can be a steadier identifier than a dynamic IPv4 one. The device half is the part that moves: privacy extensions (RFC 8981) generate a temporary identifier and rotate it, so the same device shows different addresses over time on the same network.
Neither version hides the network you are on. Geolocation reads IPv6 much as it reads IPv4, sometimes more accurately, because the blocks were allocated recently and are documented better, and what is my IP location covers what a lookup can and cannot see either way.
Frequently Asked Questions
Why does my device show both an IPv4 and an IPv6 address?
That is called dual-stack, and it is the normal setup for most modern devices and providers. Your device holds both address types at once and uses whichever one the site or service you are connecting to actually supports.
Is IPv6 faster than IPv4?
Not inherently. Any difference in practice comes from how well a particular provider or route has implemented IPv6, not from the protocol. IPv6 can skip an extra NAT translation step, which saves a negligible amount of latency, so it is not a meaningful factor for everyday use.
Can I just turn off IPv6 if I don't think I need it?
Most operating systems and routers let you, but it is rarely worth doing. It will not speed anything up, and it can add delays on networks and apps that try IPv6 first, since disabling it leaves them waiting for a connection that never answers before they fall back.
Will websites eventually stop supporting IPv4 entirely?
Eventually in theory, but not soon. Too much existing infrastructure and too many devices still depend on IPv4 for anyone to switch it off, and the transition is happening gradually through dual-stack support rather than as a cutover on a date.
Does my ISP decide whether I get an IPv6 address, or can I request one?
Your provider decides. IPv6 depends on their network infrastructure rather than anything you can configure or ask for, so if they have not deployed it you keep using IPv4 until they do. It is worth asking support, since some enable it per account on request.
Why does my IPv6 address keep changing when my IPv4 address stays the same?
Because the two halves of an IPv6 address behave differently. The network prefix from your provider tends to stay put, while privacy extensions on your device generate a temporary identifier for the second half and rotate it periodically. The address changes; the network it belongs to usually does not.
Final Words
IPv4 and IPv6 are not rivals you have to pick between. They are two systems running side by side while the internet moves from one to the other, slowly and without asking anyone. For nearly everyone the practical answer is that your device already handles it, and seeing one or both on a checker means it is working as intended. To see what your connection presents right now, read it on the MyIPShow homepage.