Validity, country, E.164 format, line type and the carrier the number block was allocated to. No account needed.
25 free checks a day from this address.
We will email a trial key for the desktop app — 50,000 numbers, offline, no card. It also raises this page from 25 checks a day to 100. Nothing else is sent to you.
valid
The number is syntactically possible for its country and its block is allocated in the numbering plan. It does not mean the line is connected today.
country / cc
Derived from the dialling code by longest-prefix match, so shared codes like +1 resolve to the right territory rather than just "North America".
e164
The number normalised to international format. This is the form most senders and CRMs want, and the form that deduplicates correctly.
line_type
Mobile, landline, VoIP, or special services, from the numbering plan record for that block. Wholesale VoIP blocks that the source labels as landline are corrected using carrier evidence.
carrier
The carrier the block was originally allocated to, shown under its trading name rather than its legal entity. Read the limits below before relying on it.
reason
Why a number failed. unmatched_prefix means no allocated block matches it,
which is the common case for a made-up or mistyped number.
Portability is the one that catches people out. A number keeps its original block assignment when it moves network, so the carrier here is the allocation, not a live query. Nothing that avoids a live network lookup can tell you otherwise, including every other free checker.
This is the single thing people get wrong about every carrier lookup, including this one, and it is worth understanding before you act on a result.
Number portability means you can change network and keep your number. In the US it has been a legal right since 2003, and most countries now have some form of it. When a number ports, the block it was allocated from does not change. Its record in the numbering plan still names the carrier the block was issued to, years after the subscriber moved.
So a number allocated to a block issued to Verizon, ported to T-Mobile in 2019, still reads as Verizon in any lookup built on allocation data. The number is correct. The carrier is historical.
From allocation data alone, you cannot. There is no flag in the numbering plan that says "this one moved". Finding out requires a live query against the routing databases carriers use to route calls — in North America the NPAC, elsewhere a national equivalent. That is a paid, per-number lookup through a provider such as Twilio or Telnyx, and it is the only thing that returns a current carrier rather than an original one.
If a free tool claims to give you the current carrier on a ported number, it is either paying for that query on your behalf or it is guessing.
Less than the caveat suggests, for most uses. Porting rates vary by market, but the allocation record is right for the large majority of numbers, and the field people usually care about — whether the number is mobile, landline or VoIP — survives porting far better than the carrier name does, because numbers rarely move between line types.
If you are choosing between SMS and voice, the line type is the field to trust. If you are billing per-carrier or routing by network, you need a live lookup.
Reachable by SMS. In countries with separate mobile ranges — most of Europe, most of Asia — line type is a property of the number range and is reliable. In the US and Canada mobile and landline share the same ranges, which is why a US number often returns a type derived from the carrier that holds the block rather than from the range itself.
Texting a landline fails, every time, and the failure is usually silent — the message is accepted by your platform and never delivered. On a list of any size this is the single largest source of wasted spend in an SMS campaign, and it is knowable before you send.
A number issued by an internet telephony provider rather than a traditional carrier. VoIP numbers can receive SMS, but they are also where disposable and short-lived numbers concentrate, so many platforms treat them as higher risk. Whether that matters depends entirely on what you are doing: for a support callback list, VoIP is fine; for fraud screening, it is a signal.
Wholesale VoIP blocks are frequently recorded as landline in the underlying data. This lookup corrects those using carrier evidence, which is why a number your other tool calls a landline may come back here as VoIP.
Not from a reseller API. The numbering plan data is assembled from the published allocation records that national regulators and numbering administrators maintain — which carrier holds which block, and what kind of service that block is for.
For the US that means NPA-NXX: the area code plus the next three digits identify a block of 10,000 numbers, and since number pooling most blocks are subdivided again into thousands-blocks of 1,000. A US lookup here matches at both depths, so it can distinguish two numbers that share an NPA-NXX but sit in differently-allocated thousands-blocks. How a US number is actually allocated covers the structure in full.
The same table holds 1.88 million blocks across 206 country codes. For +44 the UK numbering plan is authoritative and overrides the general source.
Every US NPA-NXX block is assigned to a rate centre — a geographic calling area defined for billing, not for postal geography. Rate centres are what determine whether a call between two numbers is local or toll, and they are why an area code can span several and a single city can contain many.
This lookup returns the carrier and line type, not the rate centre, because the two answer different questions. Rate centre matters when you are choosing which number to buy — provisioning a local presence, or minimising interconnection cost. Carrier and line type matter when you are deciding whether to contact a number you already have. If you need rate centre or OCN data, the Local Calling Guide publishes it free.
Run the same number through several of these tools and you will get disagreements. That is not one of them being broken. It is the honest state of the data, and almost nobody says so.
We measured it. In a pre-registered, out-of-sample benchmark in September 2026 we put the same US numbers through this engine and two independent commercial references, Twilio Lookup and Veriphone, and compared all three:
Twilio vs Veriphone
73% agreement on line type (n=513). Two paid APIs disagree with each other about a quarter of US numbers.
Phonelint vs Twilio
75% agreement (n=513).
Phonelint vs Veriphone (cached)
94% agreement (n=582).
Mobile detection
96% recall against both references; precision 97% against Twilio and 91% against Veriphone.
The figure worth sitting with is the first one. The two commercial references agree with each other only 73% of the time, so there is no scoreboard here that anyone is topping — there is a question about the US numbering plan that the industry cannot currently answer with one voice. Any tool quoting you a single accuracy percentage against "the truth" is quoting agreement with one vendor and calling it truth.
Outside North America the disagreement largely disappears, for the reason in the next section, and why libphonenumber returns FIXED_LINE_OR_MOBILE is the same limit seen from inside the code. The full benchmark has the method, the sample and the cases we got wrong.
Most free carrier lookups are US-only, and several say so on the page. This one covers 206 country codes, and the answers outside North America are usually firmer rather than weaker.
The reason is structural. In most of the world, mobile and fixed numbers are issued from separate ranges. A UK number starting 07 is a mobile; one starting 020 is a London landline. The line type is a property of the number itself, so reading it does not depend on knowing which company holds the block. In North America the two share the same ranges, so the only route to a line type is the carrier holding the block — which is precisely where the 27% disagreement above comes from.
What travels less well is the carrier name. Allocation records are published by each national regulator in its own format and on its own schedule, so a smaller market may name the licensee rather than the brand a subscriber would recognise. For +44 the UK numbering plan is authoritative here and overrides the general source.
A free lookup is the right tool right up to the point where you need one of two things: an answer for a number that has been ported, or an answer for thousands of numbers at once.
For the first, you need a live query against the routing database, and that is a paid product everywhere. Prices read off each vendor's own page in October 2026:
Veriphone
Numbering-plan validation from $0.25 per 1,000; live current-carrier lookup is a separate product at $2.50 per 1,000.
Twilio Lookup
$0.008 per request for Line Type Intelligence (checked September 2026).
ClearoutPhone
Subscription, from $23.16 a month up to $4,600 a month.
For the second, the economics change shape. A metered lookup is cheap once and expensive on repeat: at $0.008 a request, one pass over a million numbers is about $8,000, and the same pass next quarter costs it again. The desktop app does the allocation check offline at around 47,000 numbers a second on a mid-range Windows laptop, for $29 once and no per-number fee on any run after the first.
The sensible order is both. Strip the numbers that cannot ring locally, then spend live lookups only on the survivors that actually matter — which is a far smaller bill than sending the whole list to a meter.
It is sent to our API, matched against the numbering table, and answered. It is not stored against you, not added to a list, and not sold. There is no account, no sign-up and no email gate: every result on this page is free to anyone, and the email box under a result is an optional offer you can ignore for ever.
If you would rather nothing left your browser at all, the free bulk cleaner runs entirely client-side — it normalises, deduplicates and flags invalid numbers without a single network request. It cannot return a carrier, because that needs the lookup table, and that is the honest trade.
Rate limited to 25 checks a day per IP, or 100 if you have requested the free guide. That is to keep the endpoint up, not to push you toward a paid plan.
Enter it above. The result names the carrier the number's block was allocated to, under its trading name. For a number that has never been ported that is the current carrier; for one that has, it is the original. No free tool can distinguish the two without a paid live query.
This one is, with no account, up to 25 checks a day per IP. Paid APIs charge per lookup — Twilio Lookup's line-type package is $0.008 a request and Telnyx starts around $0.0015 a query (both checked September 2026) — and those fees buy a live query, which is what returns a current carrier on a ported number.
Run it above and read the line_type field. A number issued by an internet
telephony provider returns voip. Note that a VoIP number looks identical to
any other number — there is nothing in the digits themselves that marks it.
Yes, and it does not require calling it. Outside North America, mobile and fixed numbers come from separate ranges, so the number itself tells you. Inside North America they share ranges, so the answer comes from the carrier holding the block. Either way it is knowable from the number, before you send anything.
Not from this tool, and not from any tool built on allocation data. It requires a live query against the routing database — NPAC in North America — which carriers and paid APIs can make and free lookups cannot. If the distinction matters to your use case, send the survivors of a local pass to your own Twilio or Telnyx key.
No. Nothing is sent to the number, no call or message is placed, and the lookup is a database match rather than a network query. The subscriber has no way to know.
No, and this tool will never do that. It returns information about the number — its validity, format, line type and carrier allocation — not about any person. Services that claim to name a subscriber are working from leaked or scraped directories, and we are not one of them.
Accurate against what, is the question nobody asks. We benchmarked this engine against Twilio Lookup and Veriphone in September 2026 and the two commercial references agreed with each other only 73% of the time on US line type. Ours agreed 75% with Twilio and 94% with Veriphone — though every Veriphone response came back cached rather than live, which is the same method we use, so that 94% partly measures shared method rather than shared correctness. The 75% is the figure to judge us on. Invalid detection is the firm part — a number outside an allocated block cannot ring, and that is deterministic. Line type on a US number is the soft part, for everyone.
Yes. This covers 206 country codes, and outside North America the line type is usually firmer, because most countries issue mobile and fixed numbers from separate ranges. The carrier name is the part that varies: each regulator publishes allocation records in its own format, so a smaller market may return the licensee rather than the consumer brand.
One whose subscriber has moved it to a different carrier while keeping the digits. The number stays in the block its original carrier was allocated, so allocation data still names that original carrier. This is the single limitation of every free carrier lookup, including this one, and resolving it needs a live query against the routing database.
Yes. The free browser cleaner handles a whole list locally, and the desktop app does millions offline at around 47,000 numbers a second on a mid-range Windows laptop.
Clean 50,000 in your browser free, or do millions offline on your own machine with the desktop app. Neither uploads your list.