Home / IPv6 Subnet Calculator
By Ali Zafar · Last updated November 2026
IPv6 Subnet Calculator
Type an IPv6 address and a prefix length. The calculator shows the network, the shortest and the full form of the address, the last address of the subnet, how many addresses it holds, what kind of address it is, and all 128 bits with the network part marked. All maths is exact (128-bit integers) and runs in your browser.
Subnet
Binary view (128 bits, 16 per row)
1 network bits 0 device bits. The label at the start of each row is that group in hex.
How the IPv6 subnet maths works
An IPv6 address is 128 bits long, written as eight groups of four hex digits. The prefix length says how many bits from the left are the network. The calculator keeps the address as one exact 128-bit integer, so there is no rounding. The network is the address with all device bits set to 0, the last address has all device bits set to 1, and the number of addresses is 2(128 − prefix).
The shortest form of an address
RFC 5952 sets one way to write an address in its shortest form: lower case, no leading zeros in a group, and the longest run of two or more all-zero groups replaced by :: (the first run if two are equal). A single zero group is left as 0. So 2001:0db8:85a3:0000:0000:8a2e:0370:7334 becomes 2001:db8:85a3::8a2e:370:7334.
Common prefix lengths
| Prefix | Addresses | Typical use |
|---|---|---|
| /128 | 1 | One device, for example a loopback or a host route |
| /64 | 264 = 18,446,744,073,709,551,616 | One LAN segment; the usual size for a subnet |
| /56 and /48 | 256 and 65,536 subnets of /64 | Blocks that providers commonly hand to a site; check what yours gives you |
Address types the calculator recognises
| Block | Meaning | Source |
|---|---|---|
| ::/128 and ::1/128 | Unspecified and loopback | RFC 4291 |
| fe80::/10 | Link-local, valid on one link only | RFC 4291 |
| fc00::/7 | Unique local (private) addresses | RFC 4193 |
| ff00::/8 | Multicast | RFC 4291 |
| ::ffff:0:0/96 | IPv4-mapped | RFC 4291 |
| 2001:db8::/32 and 3fff::/20 | Documentation examples | RFC 3849, RFC 9637 |
| 2000::/3 | Global unicast (the internet-routable range) | RFC 4291 |
Worked example
The address 2001:db8:85a3:1234:5678:9abc:def0:1234 with prefix 64 has the network 2001:db8:85a3:1234::/64 and the last address 2001:db8:85a3:1234:ffff:ffff:ffff:ffff. It sits in the documentation block, so the calculator labels it as an example address.
Common mistakes
- Splitting a LAN below /64. Many features, such as stateless address autoconfiguration, expect a /64.
- Typing a zone id (
fe80::1%eth0) or an IPv4 ending (::ffff:192.0.2.1). Write the last 32 bits as two hex groups instead. - Counting hosts like in IPv4. There is no network or broadcast address to subtract in IPv6.
Frequently asked questions
What does the prefix length mean in IPv6?
It is the number of leading bits that identify the network. The remaining bits identify the device inside that network. A /64 leaves 64 bits for devices.
Why is a /64 the normal subnet size?
Standard IPv6 interface identifiers are 64 bits long, and features such as stateless address autoconfiguration expect a 64-bit device part. Smaller LAN subnets can break them.
How do I shorten an IPv6 address?
Drop leading zeros in each group, then replace the longest run of two or more all-zero groups with a double colon. A double colon can be used only once in an address.
Does IPv6 have a broadcast address?
No. IPv6 uses multicast instead of broadcast, so there is no broadcast address to subtract from the subnet size.
Are my addresses sent anywhere?
No. The calculation runs entirely in your browser. Nothing you type is sent to GaugeCalc or anyone else.
Related tools
For IPv4 networks use the IPv4 Subnet Calculator. To check a name's address records, use the DNS Lookup. Port numbers for camera systems are in the Ports and Protocols guide.