IPv6 Subnet Calculator
Prefix range, compressed & expanded notation, address type and reverse DNS
Accepts any valid notation: compressed (fe80::1), full, with a prefix (2001:db8::/32), a zone ID or embedded IPv4 (::ffff:192.0.2.1).
Network / interface bits
Split this prefix into smaller subnets
| # | Subnet | Last address |
|---|
IPv6 Subnetting Guide
Prefix sizes, address types and how IPv6 notation works
How to use the IPv6 subnet calculator
- Type or paste an IPv6 address. Compressed, expanded, bracketed (
[2001:db8::1]) and zone-ID forms (fe80::1%eth0) all work. - Choose the prefix length, or add it to the address as
/48,/56or/64. - Read the network prefix, first and last address, size, address type and reverse DNS zone. Use the copy icons to grab any value.
- Use Split this prefix to list the smaller subnets inside it, for example the /64s in a /56.
Common IPv6 prefix sizes
| Prefix | Typical use | /64 subnets inside |
|---|---|---|
| /32 | ISP or LIR allocation from an RIR | 4,294,967,296 |
| /48 | Enterprise site | 65,536 |
| /56 | Home or small business site | 256 |
| /60 | Small home allocation | 16 |
| /64 | Single LAN or VLAN (required for SLAAC) | 1 |
| /127 | Point-to-point router link (RFC 6164) | — |
| /128 | Single host, loopback | — |
IPv6 address types at a glance
Global unicast (2000::/3)
Publicly routable addresses assigned by your ISP or RIR. The first 48 bits usually identify the site, the next 16 the subnet and the last 64 the interface.
Unique local (fc00::/7)
Private addressing for internal networks, similar to 10.0.0.0/8 in IPv4. In practice you use fd00::/8 with a random 40-bit global ID.
Link-local (fe80::/10)
Automatically configured on every IPv6 interface and used for neighbour discovery and routing protocol next hops. Never forwarded by routers.
Multicast (ff00::/8)
IPv6 has no broadcast. Multicast groups such as ff02::1 (all nodes) and ff02::2 (all routers) replace it.
Frequently asked questions
How do I calculate an IPv6 subnet?
Keep the first n bits of the address (the prefix length) and set the remaining 128 minus n bits to zero to get the network address; set them all to one to get the last address in the prefix. A /64 always contains 2 to the power of 64 addresses. Enter any address and prefix above and the calculator does this for you.
Why is /64 the standard IPv6 subnet size?
Stateless address autoconfiguration (SLAAC) builds a host address from a 64-bit interface identifier, so LAN segments should use exactly /64. Smaller subnets break SLAAC and some neighbour discovery features. Use /127 for point-to-point router links (RFC 6164) and /128 for loopbacks.
How many /64 subnets are in a /48 or a /56?
A /48 contains 65,536 /64 subnets (2 to the power of 16) and a /56 contains 256 /64 subnets. Enterprises commonly receive a /48 per site and home users a /56 or /60 from their ISP.
How do you shorten an IPv6 address?
RFC 5952 defines the canonical form: write hex digits in lowercase, drop leading zeros in each 16-bit group, and replace the longest run of two or more all-zero groups with a double colon (::). If two runs are the same length, compress the first one. For example 2001:0db8:0000:0000:0000:ff00:0042:8329 becomes 2001:db8::ff00:42:8329.
What is the difference between fe80::/10, fc00::/7 and 2000::/3?
fe80::/10 is link-local: valid only on one network segment and never routed. fc00::/7 is unique local addressing (ULA), the IPv6 equivalent of private RFC 1918 space, normally used as fd00::/8. 2000::/3 is global unicast address space, routable on the internet.
More Networking & IP Tools
Free browser-based utilities for network engineers and system administrators
Thank you for your message.