Use the random IP address generator for test data
This random IP address generator creates address strings inside your browser. Choose IPv4 or IPv6, set a count from one to 100, and select an address pool before pressing Generate addresses. The default produces ten different IPv4 documentation addresses. Opening the page does not generate a batch, contact a generated destination, or change your connection's public address.
Read more
Start with Documentation when preparing examples, screenshots or offline test fixtures. The generator uses ranges reserved for that purpose instead of borrowing an ordinary organization's address. For a private-address parser test, choose Private / local. Select Public candidates only when your data needs addresses outside this tool's special-purpose exclusions; the label does not mean that a destination is reachable.
The generator supports a narrower range through Range and duplicates. Enter an optional CIDR such as 192.0.2.0/28. The filter intersects the selected pool, so a private prefix combined with Documentation produces an explanatory error. It does not silently switch modes or ignore your restriction. Leave the field empty to sample from the entire selected pool.
Choose a random IP address generator pool deliberately
The IPv4 documentation pool contains 192.0.2.0/24, 198.51.100.0/24 and 203.0.113.0/24: 768 literals reserved by RFC 5737 for examples. They are not live integration endpoints.
IPv6 documentation includes 2001:db8::/32 and 3fff::/20, defined by RFC 3849 and RFC 9637. Sampling is weighted by address count, so the larger block dominates. Filter to 2001:db8::/32 when you want that familiar prefix.
Private IPv4 uses the three RFC 1918 ranges: 10.0.0.0/8, 172.16.0.0/12 and 192.168.0.0/16. Private IPv6 draws from fd00::/8. Neither option detects conflicts or reserves addresses.
Understand random IP address generator results
The generator labels each batch with its family, selected pool and duplicate policy. The eligible-address count is exact, including for enormous IPv6 intervals. It counts the addresses that satisfy the pool and optional CIDR together. It is not a count of online devices, free addresses, registered owners or available connections. No scan occurs while calculating it.
Read more
Use One per line for plain-text fixtures or JSON array for quoted strings. Changing this presentation preserves the existing batch. Copy batch transfers the chosen format to the clipboard; Download batch saves the same bounded result as a text or JSON file. Neither action adds row numbers or performs destination lookups.
With No duplicates in this batch selected, the generator samples without replacement. Requesting more unique values than the eligible space contains produces an error with the exact capacity. For example, an IPv4 /30 contributes four literals when it lies entirely inside your pool. You can request all four, but not five different members of that same interval.
If repeats are allowed, the generator samples independently with replacement. A /32 IPv4 filter or /128 IPv6 filter can then produce the same single address repeatedly. Duplicate policy applies only to this operation. Separate batches may overlap, and the page does not maintain a global history or promise uniqueness across visitors, tabs or later generations.
What counts as a public candidate here?
The public-candidate policy, reviewed September 28, 2026, excludes recorded IPv4 special-purpose, multicast and reserved ranges. IPv6 starts within 2000::/3 and excludes recorded special-purpose intersections. The policy is versioned locally; generation does not fetch live IANA or BGP data.
Public candidates may be unallocated, unannounced, firewalled or already assigned to someone else. The generator does not obtain permission to use them. For an actual integration test, use a service or test endpoint you control and expect to answer. If your application normally sends traffic to every imported address, keep generated fixtures in an isolated test path with that behavior disabled.
How the random IP address generator samples large spaces
Random bytes come from Web Crypto. Missing or failed secure randomness returns an error, with no clock or Math.random fallback. A batch is shown only when complete.
Exact integers describe eligible intervals. Rejection sampling avoids uneven rank mappings, and sparse index swaps permit sampling without replacement without allocating the entire IPv6 space. Retries are bounded.
Random IP address generator privacy and limits
Settings and generated values stay in working memory, without generation requests, destination probes, URL parameters or browser storage. Ordinary page delivery and site analytics are separate network activity.
Read more
Changing generation settings removes the previous batch. Clear resets controls and results. Leaving the page discards working data; following an article section link keeps it available.
Copying deliberately transfers values out of the generator to your system clipboard. Clipboard history, synchronization or the application where you paste may retain them. Clearing this page does not erase those copies. If you need repeatable test data, save an approved fixture in your own project; the generator does not accept a deterministic seed or recover an earlier batch.