Subnet calculator

An IPv4 address is one 32-bit number written as four bytes. The prefix says how many leading bits name the network: /26 keeps 26 bits, so the mask is 255.255.255.192, the block holds 2⁶ = 64 addresses, and the address you typed is somewhere inside it. The page ANDs the address with the mask for the network, ORs it with the wildcard for the broadcast, and counts the rest. A netmask may be typed instead of a prefix (10.0.0.1 255.255.255.0), but only a contiguous one: 255.0.255.0 is refused, because no prefix means that and the result would be a guess.

Usable hosts are the block minus the network and broadcast addresses, which is why a /24 carries 254 and not 256. Two prefixes break that rule and the page follows the standards rather than the arithmetic: a /31 has no network or broadcast address — both of its two addresses are usable, which is what RFC 3021 defines for point-to-point links — and a /32 is a single host with no range at all. Leading zeros in an octet are refused too: 010 is octal 8 to some tools and decimal 10 to others, and a subnet calculator that picks one silently is worse than one that asks.

The kind line names the block the address belongs to from the registries: private ranges (10/8, 172.16/12, 192.168/16), carrier-grade NAT (100.64/10), loopback (127/8), link-local (169.254/16), multicast (224/4), the documentation ranges used in examples, and otherwise public address space. That is about the address itself, not about your network: whether it is reachable depends on routing, and no page can see that. IPv6 is a different address format and is refused here rather than half-handled.

Everything is worked out in this tab with plain 32-bit integer arithmetic. Nothing is uploaded and nothing you type is stored; the Dark button keeps its setting in this browser's local storage.