IPv6 Subnet Calculator

Type an IPv6 address and a CIDR prefix (0 to 128) and see the network, netmask, wildcard, total address count, range endpoints, and flags for private (RFC 4193), link-local, loopback, multicast, documentation (RFC 3849), and IPv4-mapped ranges. All math is BigInt so the full 128-bit range works — try ::/0. Bulk mode below for one-per-line lists. Pure client-side, no upload.

or paste
Quick fill:
Enter an IPv6 and prefix above to see the subnet breakdown.
one CIDR per line · up to 200 · # comments ok
0 lines
How IPv6 subnet math works
  • An IPv6 address is 128 bits. A CIDR prefix length N means the first N bits identify the network and the remaining (128 − N) bits identify the host.
  • The netmask is N ones followed by (128 − N) zeros. The wildcard mask is its bitwise inverse — the bits that vary inside the subnet.
  • The network address is the input IP ANDed with the mask. The last in range is the network ORed with the wildcard.
  • IPv6 has no broadcast and no reserved network/broadcast addresses the way IPv4 does. The full range is usable (RFC 4291, RFC 6164), so we just show the first and last address in the range.
  • fc00::/7 (RFC 4193) is the unique-local space, equivalent to RFC 1918 for IPv4. fe80::/10 is link-local. ff00::/8 is multicast. 2001:db8::/32 (RFC 3849) is reserved for documentation. ::1/128 is loopback. ::/128 is the unspecified address.
  • ::ffff:0:0/96 is the IPv4-mapped prefix; we detect it and show the embedded IPv4 octets. The address is not rewritten.
  • For prefix /0 the total is 2128 addresses. That's larger than JavaScript's safe integer range, so we use BigInt end-to-end and render the count with thousand-separators.
  • This tool does not parse zone identifiers (e.g. fe80::1%eth0) — they're rarely used in practice and would require a network probe to resolve.
  • Nothing leaves your browser. All math runs locally.