An IP address with a slash after it is a whole network in shorthand — and the network it names is binary arithmetic that humans are bad at doing in their heads. This CIDR Calculator expands any IPv4 address and prefix into the full picture: network and broadcast addresses, subnet and wildcard masks, the usable host range and the host count.





Worked Example: 192.168.1.10/24

The /24 prefix fixes the first 24 bits as the network portion, leaving 8 bits — 256 addresses — for hosts. The subnet mask is 24 ones followed by 8 zeros: 255.255.255.0; its mirror image, the wildcard mask, is 0.0.0.255. The network address zeroes the host bits: 192.168.1.0. The broadcast address sets them all: 192.168.1.255. Between them, 192.168.1.1 – 192.168.1.254 are assignable — 2⁸ − 2 = 254 usable hosts, the minus two being the network and broadcast addresses themselves, which name the street and shout down it rather than housing anyone. The same address at /26 shows the arithmetic biting finer: the mask becomes 255.255.255.192, the block size is 64 addresses, and 192.168.1.10 lands in the first block — network .0, broadcast .63, hosts .1–.62 (62 usable). Carve the /24 into /26s and four subnets appear (.0, .64, .128, .192), the subnetting exercise every network course assigns and every network engineer still sanity-checks with a tool exactly like this one, because the alternative is binary long division performed at 2 a.m. on a change window.

The Prefix Is the Whole Idea

CIDR — Classless Inter-Domain Routing, the 1993 reform that saved the internet from its own class system — replaced fixed Class A/B/C boundaries with the arbitrary prefix: the number after the slash is simply how many leading bits are the network’s name, and everything else follows by arithmetic. Three consequences organize all subnetting. Block sizes are powers of two: a prefix of p leaves 2^(32−p) addresses, aligned on boundaries that are multiples of the block size — a /26 can start at .0, .64, .128 or .192 and nowhere else, which is why “192.168.1.70/26” and “192.168.1.10/26” name the same subnet (.64’s block): the address inside a CIDR notation is a member, the prefix alone draws the map. Aggregation runs the film backwards: adjacent blocks sharing a prefix summarize into a shorter one (two /24s that are siblings become a /23 in a routing table), which is how the global routing system carries hundreds of thousands of networks in a table that fits in a router. The edges are special: a /31 leaves zero “usable” hosts by the minus-two rule yet is standardized for point-to-point links (RFC 3021), where both addresses work because there is no broadcast on a two-party line; a /32 is a single host route; a /0 is the entire IPv4 space — the default route, 4.3 billion addresses, the largest subnet that exists. The calculator handles the general rule and flags the edges rather than printing negative hosts at you.

Subnetting in Practice: Carving With Intent

Real subnetting is the arithmetic aimed at a design: given a block and a set of rooms, hand each room a prefix that fits its population with headroom. The method that survives contact with change windows is largest-first allocation: sort the required subnets by size, assign each the smallest aligned block that fits, and the fragmentation stays minimal — the same bin-packing instinct as loading a truck heavy-end first. A worked carve: split 10.0.0.0/24 among a 100-seat office, a 50-seat lab, a 20-seat classroom and a point-to-point link. The office needs 100 hosts → a /25 (126 usable) takes 10.0.0.0/25. The lab needs 50 → a /26 (62) takes 10.0.0.128/26. The classroom needs 20 → a /27 (30) takes 10.0.0.192/27. The link takes a /31 at 10.0.0.224/31 — and 10.0.0.226 through .255 remain free for the future, aligned and unallocated, which is what a healthy address plan looks like. Every figure in that paragraph is one run of this calculator per room. The complementary planning habit: document the allocation in the same notation the calculator speaks, because the next engineer — at the 2 a.m. change window this introduction imagined — deserves a map that the arithmetic can verify.

The Private Ranges: Three Blocks Worth Memorizing

Most CIDR arithmetic in daily life happens inside the three RFC 1918 private ranges, and the calculator prices each of them. 10.0.0.0/8 — one class-A-sized block, 16,777,214 usable addresses, the enterprise favourite because its /8 swallows any carving plan whole: a multinational’s every office is a /16 inside it with room to spare. 172.16.0.0/12 — the forgotten middle child, 1,048,574 usable addresses spanning 172.16–172.31, common in cloud defaults and VPN pools precisely because it is less likely to collide with a home router’s habits. 192.168.0.0/16 — 65,534 usable addresses, the home and small-office standard, usually served as a stack of /24s (192.168.0.0/24, 192.168.1.0/24…) by the router already sitting in your hall. The practical consequence of memorizing the three: any address beginning 10, 172.16–31, or 192.168 is private — unroutable on the public internet, free to use, and identical in millions of networks, which is both the design’s genius and the reason two merging companies (or a VPN joining home to office) discover address collisions: private space is abundant but not unique, and the carve plans of the earlier section are how large networks keep their corner of it collision-free.

Summarization: The /23 and the Routing Table’s Diet

Run the carving in reverse and CIDR pays its second dividend. Two adjacent /24s that share a parent — 192.168.0.0/24 and 192.168.1.0/24 — summarize into a single /23: mask 255.255.254.0, 512 addresses, 510 usable, network 192.168.0.0, broadcast 192.168.1.255. A routing table carrying the /23 advertises both networks in one line, and the global table stays survivable because this trick composes: a provider’s thousands of customer /24s collapse into a handful of /16s and /12s at the border. The alignment discipline from the carving section is what makes it legal — only true siblings summarize (0 and 1 pair up; 1 and 2 do not, because a /23 must start on an even boundary) — and a design that allocates sibling blocks to sibling purposes (two labs that will one day share a route, two floors of one building) is designing its future routing table at allocation time. The calculator verifies each side of the trade: run the /23 for the aggregate’s range, run the /24s for the members, and the inclusion should be exact. Address planning, done in this notation, is one of the few networking crafts where a pocket calculator and clear handwriting outperform most automation sold for the purpose.

The Small End of the Table: /29 and /30

Two prefixes do an outsized share of real-world work and deserve their own memory slot. A /29 (mask 255.255.255.248) holds 8 addresses and 6 usable hosts — the classic small-office internet block, where the provider’s gateway takes one, your firewall one, and a handful of public services share the rest. A /30 (mask 255.255.255.252) holds 4 addresses and exactly 2 usable hosts — for decades the standard point-to-point WAN block, one address for each end of the link, before /31 reclaimed the waste. Read the three smallest working sizes as a ladder of intent: /28 (14 usable) for a small server rack, /29 for the office edge, /30 or /31 for links. Each is one run of this calculator, and the pattern to notice is the tax rate: at /30, fully half the block is network and broadcast overhead — the price of the classic rule at small scales, and the entire argument RFC 3021’s /31 was written to refund on links.

Common Mistakes

Believing the address in a CIDR pair defines the subnet (the prefix does; any member address names the same block). Forgetting the minus two (or applying it to /31 links where it does not belong). Choosing a subnet that starts off-boundary — blocks must align to multiples of their size. Confusing the wildcard mask with the subnet mask (they are bitwise inverses; access lists often want the wildcard). And planning subnets exactly to today’s headcount: address space inside a private block is free, and the /26 you carve perfectly today is the renumbering project of next year — size for growth.

Related Tools

Continue with the Upload Time Calculator, the Parity Bit Calculator, the RAID Calculator, or browse the full Other Calculators hub.

Frequently Asked Questions

How many hosts are in a /24?

256 addresses total, 254 usable: 2^(32−24) = 256, minus the network and broadcast addresses. A /25 has 126 usable, a /26 has 62, a /27 has 30 and a /28 has 14 — each extra prefix bit halves the block.

What is the subnet mask for /26?

255.255.255.192. The last octet carries 2 network bits (128 + 64 = 192), the block size is 64 addresses, and subnets begin at .0, .64, .128 and .192 within the parent /24.

What is a wildcard mask?

The bitwise inverse of the subnet mask: /24’s mask 255.255.255.0 has wildcard 0.0.0.255. Router access-control lists use wildcards to say “these bits do not matter” — the ones mark the host portion.

What do /31 and /32 mean?

A /32 is a single address — a host route. A /31 is a two-address block standardized (RFC 3021) for point-to-point links, where both addresses are usable because a two-party link needs no network or broadcast address.

Can a subnet start at any address?

No — blocks align to multiples of their own size. A /26 (64 addresses) can only start at .0, .64, .128 or .192 in the last octet. That alignment is what makes routing aggregation possible.

What is the difference between network and broadcast addresses?

The network address (host bits all zero) names the subnet itself and appears in routing tables. The broadcast address (host bits all one) reaches every host on the subnet. Neither is assignable to a device in the classic rule — hence “minus two” in the host count.