Home
Enter an IPv6 address with a prefix (0–128), e.g. 2001:db8::/48, 2001:db8::1/64, or fe80::/64. If no prefix is given, /128 is assumed.
Quick prefixes:

IPv6 Common Prefix Reference

Frequently used IPv6 prefix lengths. /48 is the typical site allocation, /56 is a small site, and /64 is the standard LAN subnet. The calculator above handles every prefix from /0 to /128.

Prefix Host Bits Total Addresses /64 Subnets Typical Use

What Is an IPv6 Subnet?

An IPv6 subnet is a contiguous block of IPv6 addresses identified by a prefix. IPv6 addresses are 128 bits long (compared to 32 bits for IPv4), and the prefix length — written as /N — specifies how many of those bits are used for the network portion. The remaining 128 − N bits are host bits.

For example, 2001:db8::/48 means the first 48 bits are the network prefix and the remaining 80 bits identify hosts within that prefix. That single /48 contains 280 addresses — about 1.2 × 1024 — enough to assign a /64 to every household on Earth thousands of times over.

How to Calculate an IPv6 Network Address

The network address is found by performing a bitwise AND between the IPv6 address and the subnet mask. This zeroes out all host bits:

Example — 2001:db8:abcd:1234::1/64

Address: 2001:0db8:abcd:1234:0000:0000:0000:0001

Mask:    ffff:ffff:ffff:ffff:0000:0000:0000:0000

AND:     2001:0db8:abcd:1234:0000:0000:0000:0000

Network: 2001:db8:abcd:1234::/64

Why Is /64 the Standard IPv6 Subnet Size?

A /64 provides 264 (about 18.4 quintillion) addresses per subnet, which is far more than any single LAN could ever use. But size isn't the real reason /64 is standard — it's a protocol requirement:

  • SLAAC (Stateless Address Autoconfiguration) requires a 64-bit interface identifier, so the subnet prefix must be /64.
  • Privacy extensions (RFC 4941) and many IPv6 protocols assume a 64-bit host portion.
  • RFC 4291 recommends /64 for all LAN subnets.

Allocating a shorter prefix (e.g., /56 or /48) to a site and then carving it into /64s for each LAN is standard IPv6 address planning practice.

How Many /64s Are in a /48 or /56?

  • /48 contains 216 = 65,536 /64 subnets.
  • /56 contains 28 = 256 /64 subnets.
  • /32 contains 232 ≈ 4.29 billion /64 subnets.
  • /64 contains 1 /64 subnet (itself).

The calculator above shows the /48, /56, and /64 subnet counts automatically for any prefix of /64 or shorter, so you can quickly see how many LANs a given allocation supports.

Does IPv6 Have a Broadcast Address?

No. IPv6 replaces broadcast with multicast. The all-nodes multicast address ff02::1 serves a similar purpose, but only hosts that have joined the group receive the traffic. Within a subnet:

  • The all-zeros address (e.g., 2001:db8::) is the Subnet-Router anycast address. It must be assigned to the router interface for that subnet.
  • The all-ones address (e.g., 2001:db8:0:ffff:ffff:ffff:ffff:ffff) is a normal unicast address — no special meaning.
  • Every other address in the subnet can be freely assigned to hosts.

Because of this, there's no "usable host count" subtraction for IPv6 the way there is for IPv4. The total address count shown by this calculator is the full number of assignable addresses in the subnet.

Why Use This IPv6 Subnet Calculator?

  • 128-bit precision: All calculations use JavaScript BigInt, so counts up to 2128 are exact.
  • Complete range: Network address, last address, total addresses, and host bits.
  • Subnet planning: See how many /48, /56, and /64 subnets fit inside the prefix.
  • Multiple formats: Compressed (RFC 5952), expanded, and binary representations.
  • Full input support: Handles :: compression, IPv4-mapped addresses, and optional prefixes.
  • Free & private: Everything runs in your browser — no data is sent to a server.

❓ IPv6 Subnet Calculator FAQ

Why is /64 the standard IPv6 subnet size?

SLAAC (Stateless Address Autoconfiguration) requires a 64-bit interface identifier, so every LAN subnet must be /64. RFC 4291 recommends /64 for all LANs. A /64 provides 264 (about 18.4 quintillion) addresses, which is more than enough for any LAN.

Does IPv6 have a broadcast address?

No. IPv6 uses multicast instead of broadcast. The all-nodes multicast address ff02::1 plays a similar role, but only hosts that have joined the group receive the traffic. Within a subnet, every address except the Subnet-Router anycast is available for hosts.

What is the Subnet-Router anycast address?

It's the all-zeros address in a subnet (e.g., 2001:db8:1:2::). It must be assigned to the router interface for that subnet. It's a real address assigned to a device, not a reserved identifier like IPv4's network address.

What's the difference between /48, /56, and /64?

/48 is typically assigned to a site (a home or business) and contains 65,536 /64 LANs. /56 is a smaller allocation often given to home users and contains 256 /64 LANs. /64 is the standard LAN subnet size. An ISP typically receives a /32 from its regional registry.

How many addresses are in a /64?

264 = 18,446,744,073,709,551,616 addresses — about 18.4 quintillion. That's why a single /64 is enough for any LAN.

Can I enter a full address like 2001:db8::1/64?

Yes. The calculator will compute the network prefix (2001:db8::/64) and the full address range of that subnet. If you omit the prefix entirely, the calculator assumes /128 (a single host).

What is the largest IPv6 subnet?

/0 covers the entire IPv6 address space — 2128 addresses (≈ 3.4 × 1038). At the other extreme, /128 is a single host route.

Can I use this calculator for IPv4?

No — this tool is IPv6-only. IPv4 addresses are 32 bits long and use prefixes from /0 to /32. Use our IP Address Range Calculator or CIDR Notation Converter for IPv4.

Why are some address counts shown in scientific notation?

IPv6 totals can reach 2128 (about 3.4 × 1038), which is too large to display readably as a full number in a table cell. Scientific notation gives a compact, readable value. The exact count is still shown in the main result grid with commas.

What about IPv4-mapped addresses like ::ffff:192.168.1.1?

These are supported. The calculator converts the embedded IPv4 portion to its hexadecimal form (::ffff:c0a8:0101) and computes the subnet range in IPv6 terms. Note that a /96 within the ::ffff:0:0/96 prefix maps to the entire IPv4 space, so mixed forms work best with prefix lengths of /97 or longer.

Why does the calculator use BigInt?

JavaScript's regular Number type can only represent integers up to 253 exactly. IPv6 arithmetic requires 128-bit integers, so this calculator uses BigInt throughout. That's why the results are exact even for prefixes like /0.

Is this IPv6 subnet calculator free?

Yes. This tool is completely free, requires no registration, and runs entirely in your browser. Your inputs are never sent to a server.