Skip to main content

IPv6 Subnet Calculator

Calculate subnets, address ranges and reverse DNS for IPv6

Enter an IPv6 address with a CIDR prefix and instantly get the address range, address forms, ip6.arpa reverse DNS and address type.

Key features

Instant calculation

Get network address, last address, total IPs (2^N notation) and prefix details immediately.

Multiple address forms

View compressed, expanded, full and reverse DNS (ip6.arpa.) forms of any IPv6 address.

Address type detection

Automatically identifies Global Unicast, Link-Local, ULA, Multicast, Loopback and other types.

Smart diagnostics

Contextual info and warnings: non-standard subnets, deprecated ranges, IPv4-mapped addresses.

Free to use

The normalized network preview is visible without an account. Create a free account to unlock the full breakdown, with no usage limits.

Why use this IPv6 subnet calculator?

IPv6 addresses are 128 bits long and written in hexadecimal. Calculating subnets by hand is error-prone and slow. This IPv6 subnet calculator does it in one click: enter a CIDR block, get every detail instantly.

Use cases:

  • Network planning: size /48, /56 or /64 subnets before deploying infrastructure
  • Reverse DNS: generate the exact ip6.arpa entry for PTR record configuration
  • Address verification: confirm an IPv6 prefix is valid before pushing it to production
  • Documentation: copy the expanded, compressed and full forms for addressing plans

What is an IPv6 address?

An IPv6 address is a 128-bit identifier defined by RFC 4291. It is written as 8 groups of 4 hexadecimal digits separated by colons. The :: shorthand compresses consecutive groups of zeros into a single token.

Example:

Compressed:  2001:db8::1
Expanded:    2001:0db8:0000:0000:0000:0000:0000:0001

The CIDR prefix (e.g. /64) sets the boundary between network bits and host bits. A /64 prefix means 64 bits for the network, 64 bits for hosts.


How to calculate an IPv6 subnet

IPv6 subnetting always starts with the CIDR prefix length. The formula is straightforward:

  1. Identify the prefix: the first N bits define the network portion
  2. Calculate host bits: 128 minus N equals the bits available for hosts
  3. Determine the range: network address (all host bits = 0) to last address (all host bits = 1)

Example: 2001:db8:abcd::/48

ElementValue
Prefix/48 (48 network bits)
Host bits80 bits
Total addresses2^80 (1,208,925,819,614,629,174,706,176)
Possible /64 subnets65,536

Enter 2001:db8:abcd::/48 in the calculator above to verify these values.


Common IPv6 prefix lengths at a glance

IANA and Regional Internet Registries (RIRs like RIPE, ARIN, APNIC) assign IPv6 blocks following these standard sizes:

PrefixTypical useAddresses/64 subnets
/128Single host (loopback, host route)10
/127Point-to-point link (RFC 6164)20
/64Standard LAN subnet (RFC 4291)2^641
/56Residential customer allocation2^72256
/48Site allocation2^8065,536
/32ISP allocation from RIR2^964.3 billion

The /64 boundary is not arbitrary. SLAAC (Stateless Address Autoconfiguration, RFC 4862) requires exactly 64 host bits to function.


Key differences between IPv4 and IPv6 subnets

IPv6 eliminates several IPv4 constraints. There is no broadcast address, no subnet mask in dotted-decimal, and no need to subtract 2 from the address count.

FeatureIPv4IPv6
Address size32 bits (4.3 billion addresses)128 bits (3.4 x 10^38 addresses)
Mask notationDotted decimal or CIDRCIDR only (/N)
BroadcastYes (last address reserved)No (replaced by multicast, RFC 4291)
Standard subnet/24 (254 usable hosts)/64 (2^64 hosts)
Usable addressesTotal minus 2 (network + broadcast)All addresses in the block
Auto-configurationDHCP onlySLAAC (RFC 4862) or DHCPv6

IPv6 address types

This calculator detects the address type automatically based on the prefix, as defined in RFC 4291 and RFC 4193:

TypePrefixUse
Global Unicast2000::/3Publicly routable addresses (internet traffic)
Link-Localfe80::/10Communication within a single network segment
Unique Local (ULA)fc00::/7Private addresses, not routed on the internet (RFC 4193)
Multicastff00::/8One-to-many delivery to a group of nodes
Loopback::1/128Local interface only (equivalent of 127.0.0.1)
IPv4-mapped::ffff:0:0/96IPv4 address embedded in IPv6 format

If the calculator flags an address as "deprecated" or "reserved," check the IANA IPv6 Special-Purpose Address Registry for current status.


Generating ip6.arpa reverse DNS

Every IPv6 address has a reverse DNS (PTR) name that lives in the ip6.arpa domain. This is what lets a resolver map an address back to a hostname.

The construction rule is mechanical. Take the fully expanded address and split it into its 32 hexadecimal nibbles. Reverse their order, separate each nibble with a dot, then append the ip6.arpa suffix. This mirrors IPv4, which uses in-addr.arpa instead, but with one nibble per label rather than one octet.

Example: 2001:db8::1 becomes:

1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.8.b.d.0.1.0.0.2.ip6.arpa

Rather than expanding and reversing all 32 nibbles by hand, this IPv6 subnet calculator generates the full ip6.arpa entry for you, ready to paste into your PTR records.


FAQ: frequently asked questions

Q: What is an IPv6 subnet?

A: An IPv6 subnet is a logical division of the 128-bit address space. The standard size is /64 (2^64 addresses per LAN), as recommended by RFC 4291.


Q: What prefix length should I use?

A: /64 is the standard for LANs (required for SLAAC). /48 for a site, /56 for a residential customer, /32 for an ISP allocation, /128 for a single host.


Q: Does IPv6 have broadcast addresses?

A: No. IPv6 replaces broadcast with multicast (ff02::1 for all nodes). The last address in an IPv6 subnet is therefore usable, unlike in IPv4.


Q: What is reverse DNS for IPv6?

A: IPv6 reverse DNS uses the ip6.arpa domain. Each nibble (4 bits) of the address is reversed and separated by dots. This calculator generates the entry automatically.


Q: What happens if I don't specify a prefix?

A: The tool defaults to /64, the standard subnet size recommended by RFC 4291 for all LANs.


Q: What is an IPv4-mapped IPv6 address?

A: An address in the ::ffff:0:0/96 range (e.g. ::ffff:192.0.2.1) represents an IPv4 address embedded in IPv6 format, used during the transition to IPv6.


Q: How does CIDR notation work in IPv6?

A: CIDR notation indicates the number of network bits. 2001:db8::/32 means the first 32 bits identify the network, and the remaining 96 bits are available for subnets and hosts.


Complementary tools

ToolPurpose
WHOIS IPLook up ownership, network range and abuse contact for an IP address
Reverse DNSFind the hostname associated with an IP address
IP blacklist checkCheck whether an IP is listed on major DNSBLs
IPv4 netmaskCalculate network, broadcast and wildcard for an IPv4 address

Useful resources