Subnet Calculator

A subnet calculator for IPv4 and IPv6

This subnet calculator turns an IP address and a CIDR prefix — or a dotted subnet mask — into everything you need to plan, document or troubleshoot a network. Enter an address such as 192.168.1.10/24 and it instantly returns the network address, broadcast address, usable host range, total and usable host counts, the subnet mask and wildcard mask, the IP class, and whether the address is public, private or reserved.

It works for both IPv4 and IPv6, splits a network into smaller subnets, and shows the binary working behind every result. There's nothing to install and no sign-up — it runs entirely in your browser.

From CIDR or subnet mask to every address

Drag the prefix slider from /0 to /32 or pick a dotted mask like 255.255.255.0 — the two stay in sync, so the tool doubles as a CIDR-to-subnet-mask and subnet-mask-to-CIDR converter. For a /24 you'll see 256 total addresses, 254 usable hosts, a network of 192.168.1.0, a broadcast of 192.168.1.255 and a usable range of 192.168.1.1 – 192.168.1.254.

Special cases are handled correctly: a /31 exposes both addresses as a point-to-point link (RFC 3021) and a /32 describes a single host. Every field is one click to copy, and a 'Copy all' button grabs a tidy summary for tickets and runbooks.

See the binary — subnetting step by step

Subnetting clicks into place when you can see the bits. The visual binary view lines up the address, the subnet mask and the resulting network in binary, splitting each octet at the prefix boundary so the network bits and host bits are obvious. It's the bitwise AND that routers perform, shown step by step — ideal for students working toward CCNA, Network+ or any network engineering certification.

A compact address-map bar shows where your specific IP sits between the network and broadcast addresses, so the structure of the block is clear at a glance.

Split a network with VLSM

Cloud platforms hand you a CIDR block and expect you to carve it into subnets. The subnet splitter does exactly that: choose a longer prefix and the tool divides the network into equal blocks — a /24 into four /26s, eight /27s and so on — or enter the number of hosts each subnet needs and it picks the smallest prefix that fits (variable-length subnet masking, or VLSM).

Each resulting subnet is listed with its network address, usable host range, broadcast address and host count — the same plan you'd build for an AWS VPC, Azure Virtual Network, or GCP subnetwork.

How it compares to other subnet calculators

ipcalc (jodies.de) is the classic command-line reference tool, but it requires a terminal or copying commands into a web form with no VLSM splitting or visual binary alignment. Tunnelsup's subnet calculator is popular and simple but limited to IPv4, no VLSM, no binary view. SolarWinds' free subnet calculator is a Windows desktop app requiring download and install.

UtiloKit's subnet calculator runs directly in your browser on any device — desktop, iPhone, Android or tablet — with no install, no account, no data upload, and no ads. It handles IPv4 and IPv6, VLSM splitting, wildcard masks and a full binary view in one place.

CIDR cheat sheet and reference ranges

A built-in CIDR cheat sheet maps every common prefix to its mask, address count, usable hosts and wildcard — so /16, /24 and /30 stop being something you have to memorise. The calculator also flags private ranges (10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16), loopback, link-local and CGNAT space, and shows the legacy IP class (A, B, C, D or E) for context.

Whether you're configuring a router or firewall, designing a VPC, or revising for an exam, the math is done precisely with bitwise operations rather than by hand.

Binary subnet math, step by step

Every subnetting result comes down to a bitwise AND. Take the address 192.168.1.100/26: convert the address to binary (11000000.10101000.00000001.01100100) and write the /26 mask below it (11111111.11111111.11111111.11000000). AND the two together — a 1 AND 1 stays 1, anything else becomes 0 — and you get 11000000.10101000.00000001.01000000, which translates back to 192.168.1.64: the network address.

To find the broadcast address, flip every mask bit to get the wildcard mask (00000000.00000000.00000000.00111111 = 0.0.0.63), then OR that wildcard with the network address. The OR forces all host bits to 1 and gives 11000000.10101000.00000001.01111111 = 192.168.1.127. The usable host range is therefore 192.168.1.65 through 192.168.1.126, with the host count calculated as 2^(32 − 26) − 2 = 64 − 2 = 62 usable hosts. The two subtracted addresses are the network address itself and the broadcast address — both reserved and unassignable to any device.

This AND/OR pair is exactly what every router performs in hardware millions of times per second. Once you can do it on paper for a /26, the same method applies to any prefix: replace the last octet's boundary with wherever the prefix lands, repeat the steps, and the network address and broadcast fall out automatically. The binary view in this calculator shows each step aligned by octet so you can follow the bit math for any address you enter.

CIDR notation and the end of classful addressing

Before 1993, IPv4 addresses were divided into rigid classes. A Class A address (first bit 0) gave an organisation 16.7 million host addresses — absurdly generous for almost everyone. Class B (first two bits 10) gave 65,534 hosts, which was too many for a medium business but too few to share. Class C (first three bits 110) gave only 254 hosts, forcing large networks to request dozens of Class C blocks. The result was a routing table explosion: by the early 1990s backbone routers were carrying over 50,000 routes and memory was running out.

Classless Inter-Domain Routing (CIDR), defined in RFC 1517–1520 in September 1993, discarded the class boundaries entirely and replaced them with the slash-prefix notation used today. Any prefix length from /0 to /32 is valid, so an ISP can assign exactly the block that fits a customer — a /22 for a campus of 1,000 hosts, a /30 for a point-to-point link — rather than forcing them into Class A, B or C pigeonholes. Route aggregation (also called supernetting) is the complementary benefit: an ISP holding 256 consecutive /24 blocks can advertise a single /16 to the global table, shrinking hundreds of routes into one. This aggregation is why the internet's routing table is around 900,000 routes today instead of the hundreds of millions it would otherwise need.

CIDR also defines how address space is delegated hierarchically. IANA allocates large blocks to the five Regional Internet Registries (ARIN for North America, RIPE NCC for Europe, APNIC for Asia-Pacific, LACNIC for Latin America, AFRINIC for Africa). Each RIR sub-allocates to ISPs in blocks typically ranging from /19 to /22, and ISPs further sub-allocate to customers. Every prefix in this hierarchy is a CIDR block, and each boundary between levels is just a longer prefix — a clean, scalable design that has carried the internet from its 1993 routing crisis to the present.

Private, public, and special-purpose IP ranges

RFC 1918 (1996) reserved three IPv4 ranges for private use: 10.0.0.0/8 (16.7 million addresses), 172.16.0.0/12 (1.05 million addresses, covering 172.16.0.0 through 172.31.255.255), and 192.168.0.0/16 (65,536 addresses). These ranges are never routed on the public internet. A home router, a corporate campus or a cloud VPC can all use the same 192.168.1.0/24 block without conflict because Network Address Translation (NAT) rewrites private source addresses to a single public IP when packets leave the private network. NAT is why a single residential IP can serve dozens of laptops, phones and smart devices simultaneously.

Beyond RFC 1918, several other ranges carry special meaning. 127.0.0.0/8 is the loopback range — any address here (most commonly 127.0.0.1) routes back to the local host without touching the network stack, used by applications to communicate with services running on the same machine. 169.254.0.0/16 is the link-local range (APIPA — Automatic Private IP Addressing): Windows and macOS assign an address from this block automatically when a DHCP server is unreachable, which is why seeing a 169.254.x.x address usually means a failed DHCP negotiation. 224.0.0.0/4 is reserved for multicast (one sender, many receivers), and 240.0.0.0/4 is officially reserved for future use.

CGNAT (Carrier-Grade NAT), defined in RFC 6598, uses the range 100.64.0.0/10 (100.64.0.0 through 100.127.255.255). As public IPv4 addresses became exhausted, ISPs began placing customers behind a second layer of NAT at the carrier level — your home router gets a 100.64.x.x address from the ISP rather than a true public IP, and the ISP's infrastructure NATes that to a shared public address. CGNAT breaks peer-to-peer applications and complicates self-hosting, which is one of the practical pressures pushing the industry toward IPv6 adoption.

Subnetting patterns and the mental-arithmetic shortcuts

Network engineers who subnet daily rarely convert to binary by hand. The fastest mental shortcut is the block size rule: subtract the interesting subnet mask octet from 256 to get the block size, then count up in multiples of that block to list the subnets. For a /26, the fourth-octet mask value is 192, so the block size is 256 − 192 = 64. Starting from 0: the subnets begin at 0, 64, 128, 192 — and the broadcast of each block is one below the next boundary (63, 127, 191, 255). For a /27 the mask octet is 224, block size is 32, giving subnets at 0, 32, 64, 96, 128, 160, 192, 224. This increment method (sometimes called the magic-number shortcut) extends to any octet: a /18 has a mask of 255.255.192.0, interesting octet is the third, block size 64, subnets at x.0.0, x.64.0, x.128.0, x.192.0.

The power-of-two table is worth memorising from /24 outward: /24 = 256 addresses (254 usable), /25 = 128 (126), /26 = 64 (62), /27 = 32 (30), /28 = 16 (14), /29 = 8 (6), /30 = 4 (2), /31 = 2 (point-to-point link, RFC 3021), /32 = 1 (host route). The /30 is the go-to for router-to-router links because it wastes only two addresses; the /29 fits a small server cluster with 6 usable hosts; the /28 suits a DMZ or management VLAN with up to 14 devices.

VLSM (Variable-Length Subnet Masking) takes this further: instead of dividing an address space into equal-sized blocks, you fit each subnet precisely to its need. A practical example — allocate 192.168.10.0/24 across five departments: Sales needs 100 hosts (/25, 126 usable, takes 192.168.10.0/25), Engineering needs 50 (/26, 62 usable, takes 192.168.10.128/26), HR needs 20 (/27, 30 usable, takes 192.168.10.192/27), Finance needs 10 (/28, 14 usable, takes 192.168.10.224/28), Management needs 5 (/29, 6 usable, takes 192.168.10.240/29). The remaining 192.168.10.248/29 stays in reserve — no address is wasted. VLSM is the technique behind efficient AWS VPC design, and the subnet splitter in this tool automates it: enter the host counts in order from largest to smallest and it assigns blocks with no overlap and no waste.

Private, instant and free

Every calculation happens locally in your browser, so the addresses and network plans you enter never leave your device and the tool keeps working offline. Bookmark this subnet calculator and reach for it whenever you need to convert a CIDR, find a host range, build a wildcard mask, or split a network — from your laptop or your phone, no account required.