Subnet IP Enumerator & CIDR Calculator

How CIDR notation works

CIDR (Classless Inter-Domain Routing) replaced the old class-based IPv4 scheme in the 1990s and is now the way every router, firewall, cloud VPC and Kubernetes networking model describes an address block. The notation a.b.c.d/N means "the address a.b.c.d with the first N bits fixed as the network identifier and the remaining 32 − N bits available for hosts". A /24 therefore reserves 24 network bits and 8 host bits, giving 256 total addresses and 254 usable hosts after you remove the network and broadcast.

Worked IPv4 examples

Reserved & private ranges to recognise

• RFC 1918 private IPv4 — 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16. Safe to use behind NAT.
• RFC 6598 CGNAT — 100.64.0.0/10. Reserved for carrier-grade NAT; sometimes shows up on mobile networks.
• Loopback — 127.0.0.0/8 (IPv4) and ::1/128 (IPv6).
• Link-local — 169.254.0.0/16 (APIPA) and fe80::/10 in IPv6 for on-link communication without a router.
• IPv6 unique local — fc00::/7 (RFC 4193). The IPv6 equivalent of RFC 1918, intended for site-local addressing.
• Multicast — 224.0.0.0/4 in IPv4 and ff00::/8 in IPv6.

IPv6 differences worth remembering

IPv6 removes the concept of a broadcast address and uses well-known multicast groups instead, so this tool reports the broadcast field as not applicable for any v6 input. Subnets are typically allocated in much larger sizes: /64 is the standard LAN, /56 a small site, and /48 a full customer allocation. Because a single /64 contains 18 quintillion addresses, enumerating one is meaningless — that is why the enumerator caps "Show all" at 65,536 addresses (equivalent to a /112). For real network design you read v6 summaries, you don't list IPs.

The /31 and /32 special cases

A /32 represents a single host route — common in BGP loopback advertisements and Kubernetes service IPs. A /30 traditionally provides two usable hosts on a point-to-point link. RFC 3021 added /31 as a more efficient point-to-point block where both addresses are usable, saving address space on WAN circuits. This tool follows RFC 3021, so a /31 reports two usable hosts rather than zero.

Common mistakes

1. Treating the broadcast as usable. In a classic IPv4 subnet (/30 through /24) the network address and broadcast must not be assigned to hosts. Use the "Include network & broadcast" toggle to see them in the list when you want to verify the boundary.
2. Forgetting host bits must be zero. 192.168.1.55/24 is a host in the 192.168.1.0/24 network — the calculator normalises it for you and shows a warning, but route configurations expect the network form.
3. Mixing dotted-decimal and CIDR. A mask of 255.255.255.0 equals /24; learn the common pairs by heart (/24=…255.0, /25=…128, /26=…192, /27=…224, /28=…240).
4. Trying to enumerate huge subnets. A /8 has 16 million addresses. Enumeration is the wrong tool — use the summary, then split into smaller children if you actually need addressable ranges.
5. Overlapping VPC ranges. If two cloud VPCs you want to peer share any address bits, peering will fail. Paste both CIDRs into the aggregator and check whether the smallest covering prefix is suspiciously small — that means they overlap.

Quick reference: prefix → mask → hosts