IPv4 and IPv6 are two versions of the protocol used to address and route internet packets. They serve the same basic purpose, but they use different packet formats and address spaces. Neither version's address can simply be relabeled as the other.
The differences at a glance
| Feature | IPv4 | IPv6 |
|---|---|---|
| Address length | 32 bits | 128 bits |
| Written format | Four decimal numbers separated by dots | Hexadecimal groups separated by colons |
| Documentation example | 192.0.2.10 | 2001:db8::10 |
| Address space | About 4.3 billion total addresses | 2128 total addresses |
| Typical home setup | Private device addresses with translation at a router | A routed prefix with separate addresses for devices |
| Private local addressing | RFC 1918 ranges | Unique-local and link-local scopes |
| Firewall needed | Yes | Yes |
The example addresses above are reserved for documentation. Total address-space figures include addresses that are reserved or unsuitable for public use.
Why IPv6 exists
IPv4's address space is small relative to the number of connected devices. Network address translation (NAT) lets many devices share a public IPv4 address, and carrier-grade NAT can share an address across multiple subscribers. These approaches conserve addresses but can complicate inbound connections and troubleshooting.
IPv6 provides a much larger address space. Providers can allocate a prefix to a customer network, allowing devices to use their own IPv6 addresses. A globally scoped address does not mean unsolicited inbound traffic is allowed: router and host firewall rules still control reachability.
Why you may have both addresses
A dual-stack network supports both protocols. If a hostname has IPv4 and IPv6 records, your browser chooses a working connection, often by attempting alternatives with a short delay between them. The address shown on this site is the address used for that particular request.
Seeing an IPv4 result therefore does not prove that your connection lacks IPv6. The browser may have selected IPv4, the destination may lack IPv6, or one path may have been slower. Likewise, an IPv6 result does not reveal your separate public IPv4 address.
What counts as a real IPv6 test?
A successful request over IPv6 demonstrates a working IPv6 path to that destination at that moment. A dedicated IPv6-only endpoint can test that path without allowing the browser to fall back to IPv4. The endpoint must actually have the appropriate DNS and network configuration.
If such an endpoint is unavailable or not configured, the honest result is “not verified.” A failed request can reflect DNS, filtering, routing, the endpoint itself, or the local network; it does not identify the cause by itself. Use the IPv6 check to see what this request can establish, and the IPv4 check for the other address family.
Is IPv6 faster or more secure?
Neither protocol guarantees a faster connection. Performance depends on the provider, routing, peering, congestion, and destination. A connection can have a better path over either protocol.
IPv6 also does not automatically make a device secure. Keep firewall policies appropriate for both protocols, update software, and use encrypted application protocols such as HTTPS. NAT is an address translation mechanism and should not be treated as a complete security policy.