A firewall export lists addresses as 32-bit integers, an application log shows ::ffff:198.51.100.7, a PTR record has to be created for 2001:db8::1, and someone pastes an address with leading zeros that two tools interpret differently. All of these are notation problems: the address is a number, and you need it in a different form.
The IP Address Converter parses IPv4 and IPv6 addresses — including IPv4 written as a decimal integer, hex or binary — and shows the exact value in every common notation: dotted decimal, integer, hex, binary, dotted octal, expanded and compressed IPv6, reverse-DNS names and IPv4-mapped forms. It validates strictly, performs no lookups and runs in your browser.
What is the IP Address Converter?
In simple terms
You paste an address; the tool shows the same address written in all the other ways software and documentation use, and tells you which kind of address it is (private, documentation, loopback, public…).
Technical background
An IPv4 address is a 32-bit unsigned number. The familiar form is four decimal octets (0–255) separated by dots. The same value can be written as one integer from 0 to 4,294,967,295, as eight hex digits, or as 32 bits.
An IPv6 address is a 128-bit number written as eight groups of 16 bits in hex, separated by colons. RFC 4291 allows leading zeros in a group to be dropped and one run of zero groups to be shortened to ::. RFC 5952 defines a single canonical text form: lowercase, no leading zeros, and :: for the longest run of at least two zero groups (the first, on a tie).
Special forms embed IPv4 in IPv6: IPv4-mapped addresses (::ffff:0:0/96) used by dual-stack sockets, and the NAT64 well-known prefix 64:ff9b::/96 from RFC 6052.
Reverse DNS names put the address into the in-addr.arpa (IPv4, octets reversed) or ip6.arpa (IPv6, nibbles reversed) tree, where PTR records live.
Why use the IP Address Converter?
- Read integer and hex addresses from databases, firewall exports, packet dissectors or malware configuration.
- Normalize IPv6 to the canonical compressed form before comparing addresses or writing ACLs, and expand it for tools that need fixed-width fields.
- Build PTR names for
ip6.arpawithout counting 32 nibbles by hand. - Spot ambiguity: leading zeros, zone IDs and prefixes are refused with a clear reason.
- Classify quickly: private (RFC 1918), CGNAT, link-local, documentation, multicast or public.
Who should use it?
Network engineers and system administrators working with ACLs, logs and DNS; developers storing addresses as integers; security analysts decoding indicators in alternative notations; DNS administrators creating reverse zones. For networks and masks rather than single addresses, use the Subnet Mask Converter or the Subnet Calculator.
How the IP Address Converter works
Everything is pure arithmetic in your browser; no DNS, WHOIS or geolocation request is made.
- Cleaning. Leading and trailing spaces are removed, and square brackets around an IPv6 address (
[2001:db8::1]) are stripped. - Refusals. A zone index (
fe80::1%eth0) is rejected — it is local to one host and not part of the address. A prefix length (10.0.0.0/8) is rejected with a pointer to the Subnet Mask Converter. - IPv4 dotted decimal. Four octets 0–255 without leading zeros.
010.0.0.1or256.1.1.1produce an error explaining the rule. - IPv4 binary.
0bfollowed by up to 32 binary digits, exactly 32 binary digits, or four dotted groups of 8 bits. - IPv6. Colon notation with at most one
::, groups of 1–4 hex digits, optionally ending in an embedded dotted IPv4 (::ffff:192.0.2.1,64:ff9b::192.0.2.33). - IPv4 integer and hex. A bare decimal integer up to 4,294,967,295, or
0xwith 1–8 hex digits. Larger integers are refused because they could only be IPv6, which must be written in colon notation. - Output. The 32- or 128-bit value is written in each notation. The address type comes from a table of IANA special-purpose ranges.
The ?ip= query parameter pre-fills the input, so /ip-converter?ip=192.0.2.1 opens with that address.
How to use the NetTools.tech IP Address Converter
- Open the IP Address Converter.
- Type or paste into IP address: dotted IPv4 (
192.0.2.1), IPv6 (2001:db8::1), or a 32-bit IPv4 value as decimal,0x-hex or binary. - Or click a Try: example:
192.168.1.1,3232235777,0xC0A80101,2001:db8::1,::ffff:192.0.2.1,fe80::1ff:fe23:4567:890a. - Read Representations. The line under the heading shows IPv4 or IPv6 and, for numeric input, how it was interpreted ("Decimal integer interpreted as a 32-bit IPv4 address.").
- Copy single values with their copy buttons, or export everything with CSV, JSON, Copy all or Print. Reset restores the default address.
Understanding the results
| Result | Meaning | What to look for |
|---|---|---|
| Dotted decimal (IPv4) | Standard notation | The normalized form of any numeric input |
| Decimal (32-bit integer) | Unsigned value 0–4,294,967,295 | Matches INET_ATON() and similar functions |
| Hexadecimal / Dotted hexadecimal | 0xC0000201 / C0.00.02.01 | Handy for packet dumps |
| Binary (dotted octets / 32 bits) | Bit pattern | Compare with mask bits |
| Dotted octal | Legacy form such as 0300.00.02.01 | Avoid in configuration |
| IPv4-mapped IPv6 | ::ffff:a.b.c.d, with the hex form as a note | How dual-stack logs show IPv4 clients |
| Compressed (RFC 5952) | Canonical IPv6 text form | Use for comparisons and documentation |
| Expanded | Eight 4-digit groups | Fixed-width, sortable |
| Hexadecimal (128-bit) / Binary (16-bit groups) / Decimal (128-bit integer) | Numeric forms of IPv6 | The decimal form is rarely used in practice |
| Reverse DNS name | in-addr.arpa or ip6.arpa owner name | The name a PTR record needs |
| Embedded IPv4 (mapped / NAT64) | IPv4 inside ::ffff:0:0/96 or 64:ff9b::/96 | Only shown when present |
| Historical class | Classful A–E | Reference only; CIDR replaced classes in 1993 |
| Address type | IANA special-purpose category or Public | Not an ownership or location lookup |
Example: one IPv4 address in every form
Enter 192.0.2.1 (a documentation address):
Dotted decimal 192.0.2.1
Decimal (32-bit integer) 3221225985
Hexadecimal 0xC0000201
Dotted hexadecimal C0.00.02.01
Binary (dotted octets) 11000000.00000000.00000010.00000001
Binary (32 bits) 11000000000000000000001000000001
Dotted octal 0300.00.02.01
IPv4-mapped IPv6 ::ffff:192.0.2.1
Reverse DNS name 1.2.0.192.in-addr.arpa
Historical class C
Address type Documentation (TEST-NET-1)Entering 3221225985 or 0xC0000201 produces the same list. Now enter 2001:0db8:0000:0000:0000:0000:0000:0001: the compressed form is 2001:db8::1, the 128-bit decimal value is 42540766411282592856903984951653826561, the reverse name is 1.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.8.b.d.0.1.0.0.2.ip6.arpa, and the address type is Documentation.
For ::ffff:192.0.2.1, note that the compressed line shows the pure hex form ::ffff:c000:201, while the Embedded IPv4 (mapped) line gives ::ffff:192.0.2.1.
Common use cases
- Converting integer-stored addresses from databases or SIEM exports.
- Normalizing IPv6 addresses from different log sources before deduplication.
- Creating PTR records for IPv6 hosts in an
ip6.arpazone. - Recognizing IPv4 clients behind
::ffff:notation in web server logs. - Decoding obfuscated address notations (hex, integer) in phishing URLs or malware samples.
- Teaching how dotted notation relates to the underlying bits.
Troubleshooting
"each part must be 0–255 without leading zeros"
Remove leading zeros (10.0.0.1, not 010.000.000.001) and check that no octet exceeds 255.
"Remove the zone index"
Drop the %interface suffix. It identifies an interface on the local host and is not part of the address.
"Enter a single address without a prefix length"
The converter handles single addresses. For 10.0.0.0/8, use the Subnet Mask Converter or Subnet Calculator.
"is larger than 4,294,967,295"
The integer does not fit in 32 bits. If it is an IPv6 value, write the address in colon notation instead.
"is not a valid IPv6 address"
Check for more than one ::, groups longer than four hex digits, non-hex characters, or an invalid embedded IPv4 tail.
Common mistakes
- Treating Address type as a lookup. "Public" only means the address is not in a special-purpose range; it says nothing about who uses it.
- Using dotted octal or hex in configs. Parsers differ; use dotted decimal.
- Comparing IPv6 as strings.
2001:DB8::0:1and2001:db8::1are equal; compare compressed or expanded forms. - Expecting the compressed line to show mixed notation. It always prints hex groups; embedded IPv4 is shown on its own line.
- Relying on classful addressing. Classes are shown for reference only.
Limitations
- No lookups: no owner, ASN, geolocation, reverse-DNS resolution or reachability.
- Single addresses only; networks and prefixes are handled by the subnet tools.
- IPv6 integers cannot be entered, only displayed.
- Address type uses a fixed table of special-purpose ranges and may lag behind new IANA assignments.
- Shorthand IPv4 forms (such as
127.1) and per-octet hex or octal input are deliberately not accepted.
Security & privacy
The conversion runs entirely in your browser; the address you enter is not sent to NetTools.tech or anyone else. The page may send an anonymous usage counter (tool name and view or run, no inputs) unless your browser sends Global Privacy Control or Do Not Track. If you prefill the input with ?ip=, remember that the address then appears in the URL and in your browser history.
Related tools
- Subnet Mask Converter — masks, prefixes, ranges and subnet splitting. Guide: Subnet mask converter guide.
- Subnet Calculator — IPv4 and IPv6 network details. Guide: Subnet calculator guide.
- IP Lookup — registry, network and abuse contact for public addresses. Guide: IP lookup guide.
- Reverse DNS — resolve the PTR record for an address. Guide: Reverse DNS lookup guide.
- Number Base Converter — general binary, hex and decimal conversion.
- Category overview: Conversion & calculation tools.
Summary
The IP Address Converter treats an IP address as what it is — a 32- or 128-bit number — and writes it in dotted, integer, hex, binary, canonical and expanded IPv6, reverse-DNS and IPv4-mapped forms. It validates strictly, refusing leading zeros, zone IDs and prefixes, and it never looks anything up. Use it for notation; use the lookup tools for ownership and DNS.
Frequently asked questions
How do I convert an IPv4 address to a decimal integer?
Multiply the octets by powers of 256: a × 16,777,216 + b × 65,536 + c × 256 + d. For 192.0.2.1 that is 3,221,225,985, which is what the converter shows as Decimal (32-bit integer).
Why are IPv4 addresses with leading zeros rejected?
Some software reads a leading zero as octal, so 010.0.0.1 can mean 8.0.0.1 in one tool and 10.0.0.1 in another. To avoid that ambiguity the converter refuses octets with leading zeros.
What is the canonical IPv6 form?
RFC 5952 defines it: lowercase hex, no leading zeros in a group, and the longest run of two or more zero groups replaced by :: (the first one if there is a tie). A single zero group is not compressed.
Can I enter an IPv6 address as a number?
No. A bare integer, 0x-hex value or 32-digit binary string is only accepted as a 32-bit IPv4 value, where the meaning is unambiguous. IPv6 must be written in colon notation; its 128-bit integer is shown as output.
Does the converter tell me who owns an address or where it is?
No. It performs no lookups at all, and notation says nothing about ownership, location or reachability. The Address type line only classifies the address against the IANA special-purpose ranges; use IP Lookup for registry data.
What is an IPv4-mapped IPv6 address?
It is an IPv4 address embedded in the ::ffff:0:0/96 range, such as ::ffff:192.0.2.1. Dual-stack software uses it to represent IPv4 peers on IPv6 sockets, which is why it appears in server logs.
How is the reverse DNS name built?
For IPv4 the octets are reversed and in-addr.arpa is appended, so 192.0.2.1 becomes 1.2.0.192.in-addr.arpa. For IPv6 all 32 hex nibbles are reversed, separated by dots and followed by ip6.arpa.