Classify an IPv4 address beyond “private or public”
This tool identifies whether an IPv4 address belongs to RFC 1918 private space, CGNAT shared space or several common special-use ranges. It also displays the historical A/B/C/D/E class derived from the first octet, while clearly separating that legacy classification from modern CIDR-based network design.
Use it when reviewing an unfamiliar address, reading logs or checking whether an address should appear on the public Internet. For the broader addressing model, see the IP addressing guide.
How to use the IPv4 classifier
Enter one dotted-decimal IPv4 address and press Classify IP. The result shows the historical class, a high-level type, any matching private range, any detected special-use range and useful flags such as CGNAT, loopback or multicast.
The built-in CGNAT example is useful because 100.64.0.0/10 is neither ordinary public unicast nor RFC 1918 private space. Treating it as simply “public” can cause confusion when troubleshooting carrier networks.
Private IPv4 ranges
RFC 1918 reserves three ranges for private internetworks: 10.0.0.0/8, 172.16.0.0/12 and 192.168.0.0/16. These addresses can be reused independently inside many organizations and are commonly translated with NAT when reaching the public Internet.
Private does not mean automatically secure. Reachability and security depend on routing, firewalls, ACLs and the surrounding architecture. The NAT/PAT guide explains how address translation relates to private IPv4 deployment.
CGNAT and other special-use space
The shared address block 100.64.0.0/10 is commonly used by service providers for carrier-grade NAT. Other important ranges include 127.0.0.0/8 for loopback, 169.254.0.0/16 for IPv4 link-local addressing, documentation prefixes such as 192.0.2.0/24, and 224.0.0.0/4 for multicast.
The classifier surfaces these ranges explicitly so that “not RFC 1918” is not mistaken for “ordinary public unicast.” That distinction is valuable when interpreting traceroutes, inventory exports and packet captures.
Why class A/B/C is historical
Classful networking divided IPv4 space according to the first octet and implied fixed network sizes. Modern routing and subnet design use CIDR prefixes instead. A real network may be /20, /27 or /31 regardless of the old class associated with the first octet.
The tool still displays the historical class because it appears in training material and older documentation, but you should use the actual CIDR prefix for calculations. The CIDR calculator is the appropriate tool for capacity and boundary work.
Public unicast is a residual category
If an address is not private and does not match one of the special ranges detected by this calculator, the tool labels it as ordinary public unicast. That label does not guarantee that the address is globally routed, currently allocated or reachable from your network. It only describes the classification available from the local range checks.
For real-world ownership or geolocation questions, use the dedicated diagnostic tools rather than inferring organization or location from the numeric address alone.
Frequently asked questions
Is 100.64.0.0/10 private?
It is shared address space for CGNAT, not one of the three RFC 1918 private ranges.
Are class A, B and C still used for subnet design?
No. Modern design uses CIDR prefixes. The classes are retained only as historical context.
Is every non-private IPv4 address publicly reachable?
No. Special-use space, routing policy, allocation state and firewalling can all prevent public reachability.
What is 169.254.0.0/16?
It is IPv4 link-local space, often seen when a host cannot obtain normal addressing through DHCP.
Next step
Once you know the address category, use CIDR, routing and diagnostic tools to determine the actual network boundaries and reachability instead of relying on classification alone.