# Mobile Proxy Fraud Score, Tested on 55 Exits

> Mobile proxy fraud score, tested 15 October 2026: 55 carrier exits in 11 countries, a median risk score of 0 in every country and 50 of 55 at 0. From $5.

[Home](https://quanticdata.io/)/[Blog](https://quanticdata.io/blog/)/Mobile Proxy Fraud Score, Tested on 55 Exits

# Mobile proxy fraud score, tested on 55 carrier exits in 11 countries

ProxiesOct 15, 2026·7 min read·By [Aldo Morese](https://quanticdata.io/about/), founder of QuanticData

Four QuanticData mobile exits in the United States, Germany, India and Indonesia, on carriers such as AT&T, T-Mobile, Deutsche Telekom, Vodafone, Jio and Telkomsel, looked up on proxycheck.io; the scorecard of 15 October 2026 shows 55 exits in 11 countries, a median risk score of 0 in every country, 50 of 55 at 0 and no VPN, Tor or hosting flag.

On this page [What we checked, and how](/blog/mobile-proxies-fraud-score-tested/#what-we-checked-and-how) [Results at a glance](/blog/mobile-proxies-fraud-score-tested/#results-at-a-glance) [Country by country](/blog/mobile-proxies-fraud-score-tested/#country-by-country) [Why carrier exits score low](/blog/mobile-proxies-fraud-score-tested/#why-carrier-exits-score-low) [How to check the exit you were given](/blog/mobile-proxies-fraud-score-tested/#how-to-check-the-exit-you-were-given) [Limits of this test](/blog/mobile-proxies-fraud-score-tested/#limits-of-this-test) [The setting to buy and what it costs](/blog/mobile-proxies-fraud-score-tested/#the-setting-to-buy-and-what-it-costs)

Mobile proxy fraud score, measured on 15 October 2026: we opened 55 QuanticData mobile exits in 11 countries with the carrier pinned and looked each address up off the proxy. On proxycheck.io 50 of 55 scored 0 out of 100 and the median was 0 in every country. IP2Location.io flagged none of the 55 as a proxy.

"0 fraud score" is the phrase buyers search for and sellers print in their titles, usually without saying which service gave the zero, on which address, or when. This page is the opposite: one evening, two public services, the settings and the time of every lookup, and the per-country summary in [results-2026-10-15.csv](https://quanticdata.io/blog/mobile-proxies-fraud-score-tested/results-2026-10-15.csv?v=20261015a).

## What we checked, and how

A fraud score belongs to one IP address at one moment, on one service. So the test reads the address each exit actually used, then asks the services about that address straight away, from outside the proxy.

| Item | Setting |
| --- | --- |
| Date and time | 15 October 2026, 20:13 to 20:16 UTC |
| Network | QuanticData mobile (4G/5G), rotating, one request per exit |
| Countries | United States, United Kingdom, Germany, Brazil, India, Italy, South Korea, Greece, Indonesia, Russia, Philippines |
| Exits | 5 distinct addresses per country, 55 in total, the carrier pinned in the username and taken in turn from the carriers listed on each country page |
| Exit address | read through the proxy from ipinfo.io/json, the same command as the example on our country pages |
| Reputation, service 1 | proxycheck.io API v3, default settings, no API key: risk score from 0 to 100 plus proxy, VPN, Tor, hosting and compromised detections |
| Reputation, service 2 | IP2Location.io API, no API key: proxy flag and country |
| Not published | the IP addresses themselves |

Both services were queried within seconds of the exit being read, one lookup at a time and well inside their free daily limits. We picked them because their terms allow automated lookups by a business; we did not scrape score pages whose terms forbid it.

**Try it on your own targets.** Every QuanticData account gets $2 of free API usage per month.

[Start free — $2/month included](https://quanticdata.io/signup/?utm_source=blog&utm_medium=website&utm_campaign=blog-inline&utm_content=mobile-proxies-fraud-score-tested)[Explore 4G/5G Mobile Proxies: City & ISP Targeting](https://quanticdata.io/mobile-proxies/)

## Results at a glance

| Measure | Result |
| --- | --- |
| Exits checked | 55 in 11 countries |
| Exits in the requested country (ipinfo.io, proxycheck.io and IP2Location.io agree) | 55 of 55 |
| Risk score 0 on proxycheck.io | 50 of 55 |
| Median risk score on proxycheck.io | 0 in each of the 11 countries |
| Flagged as VPN, Tor, hosting or compromised on proxycheck.io | 0 of 55 |
| Flagged as a proxy on IP2Location.io | 0 of 55 |

On proxycheck.io a risk score starts from a baseline set by what the address is used for (hosting, VPN, scraper, Tor, proxy) and rises with recent attacks seen from it. A score of 0 therefore says that, at the moment of the lookup, the address was not in any of those groups and had no attack history, which is what a carrier address shared by ordinary phones should look like.

## Country by country

| Country | Carriers pinned | Exits | Median risk score |
| --- | --- | --- | --- |
| United States | AT&T Wireless, T-Mobile, Verizon Wireless, C Spire | 5 | 0 |
| United Kingdom | BT, O2, Three | 5 | 0 |
| Germany | Deutsche Telekom, Vodafone, O2 | 5 | 0 |
| Brazil | Vivo, Claro, TIM Brasil | 5 | 0 |
| India | Jio, Airtel, Vodafone Idea, BSNL | 5 | 0 |
| Italy | TIM, Vodafone Italia, WindTre, Fastweb | 5 | 0 |
| South Korea | KT, LG Uplus, SK Telecom | 5 | 0 |
| Greece | Nova Greece, Vodafone Greece | 5 | 0 |
| Indonesia | Telkomsel, Indosat, XL Axiata, Smartfren, Orion Cyber Internet | 5 | 0 |
| Russia | MTS, Megafon, Beeline, Tele2 Russia, T-Mob | 5 | 0 |
| Philippines | Smart Communications, Globe Telecom, Globe Business Gfiber Broadband Plan, Dito Telecommunity | 5 | 0 |
| **All 11** | **40 carriers** | **55** | **0** |

Each of these 11 country pages now carries its own line from this run: the carriers we pinned, the date and the result. Eight of them also show the network (ASN) each carrier's exits came out on, measured the same day, so you can compare a carrier, its ASN and its reputation in one place.

## Why carrier exits score low

A mobile operator puts thousands of subscribers behind each public address with carrier-grade NAT. Fraud and proxy databases know this: an address that serves a whole neighbourhood of phones cannot be blocked without blocking real customers, and most of the traffic behind it is ordinary app and browser use. That is the structural reason [QuanticData mobile proxies](https://quanticdata.io/mobile-proxies/) sit at the bottom of the scale, and why datacenter ranges, which a hosting flag catches at once, sit at the top.

It is also why the result depends on the service you ask. Each checker has its own data and its own model: one weighs recent attacks, another the share of suspicious traffic it has seen from the whole carrier, another lists of known proxy endpoints. The same address can read 0 on one and higher on another, and any score can move over days. Treat a number as a snapshot of one address on one service, and write down which one.

## How to check the exit you were given

Check the address your session actually uses, not the gateway hostname, and check it on the service your target cares about. With a sticky session the address holds for the session lifetime, so one lookup covers the whole job.

```
# 1. read the exit address through the proxy
curl -x "http://USER-country-de-isp-deutsche_telekom_ag-sid-check01-ttl-30m:PASS@$PROXY_HOST" https://ipinfo.io/json

# 2. look that address up off the proxy (replace EXIT_IP)
curl "https://proxycheck.io/v3/EXIT_IP"
```

Note the service, the settings you used and the time, as we did above. If an address does not suit you, start a new session id and you get a new exit from the same carrier and country. To time and compare exits across countries without writing code, the [proxy tester](https://quanticdata.io/tools/proxy-tester/) builds the same commands in curl, Python and Node.

| Job | What to check | Why |
| --- | --- | --- |
| Account work on a social or marketplace app | The sticky exit, once per session, on the checker you rely on | The address holds for the session lifetime |
| Ad and content checks by market | Country and carrier of the exit, then the score | The view depends on the carrier and the country |
| A fixed address with a known score for weeks | A static ISP address instead of a rotating mobile one | Mobile addresses rotate by design |

If you need one address that keeps a low score for weeks, a rotating mobile pool is the wrong tool: see [low fraud score ISP proxies](https://quanticdata.io/isp-proxies/low-fraud-score/): a private static address kept for a renewable 30-day term.

## Limits of this test

- One evening, five exits per country: enough to see a pattern by country, not to promise a rate for every address.

- Two services. Other checkers use other data, and their numbers for the same address can differ.

- A score describes one address at one moment; mobile addresses rotate and scores change over time.

- We looked up reputation only. How a platform treats an account depends on much more than the address; for what two big platforms return through these exits, see [the Instagram and TikTok test of 4 October 2026](https://quanticdata.io/blog/mobile-proxies-instagram-tiktok-tested/), and for speed, sessions and geo match, [our 98-request test of the mobile network](https://quanticdata.io/blog/best-mobile-proxies-tested/).

## The setting to buy and what it costs

For the workflows above: the mobile network, the country and, if you want it, the carrier in the username, rotating or sticky, HTTP or SOCKS5. Pinning a carrier is `-isp-` plus the code shown on each country page; leaving it out rotates across every carrier of the country.

```
curl -x "http://USER-country-us-isp-t_mobile_usa:PASS@$PROXY_HOST" https://ipinfo.io/json
```

Mobile bandwidth is $5 for 1 GB, valid 90 days, down to $2.30 per GB at 500 GB, as one-off packs with no subscription. Country, carrier, region and city targeting are included. All packs are on the [pricing page](https://quanticdata.io/pricing/).

### Sources & further reading

- [proxycheck.io API documentation: v3 detections and the 0-100 risk score used for every lookup](https://proxycheck.io/api/)

- [IP2Location.io documentation: the IP geolocation API and its proxy flag](https://www.ip2location.io/ip2location-documentation)

- [IPinfo developer documentation: the /json endpoint used to read each exit address](https://ipinfo.io/developers)

- [RFC 6598: the shared address space reserved for carrier-grade NAT](https://www.rfc-editor.org/rfc/rfc6598)

- [Wikipedia: Carrier-grade NAT](https://en.wikipedia.org/wiki/Carrier-grade_NAT)

## FAQ

Quick answers on mobile proxy fraud score.

[Something else? Ask us](mailto:hello@quanticdata.io)

### Do mobile proxies have a low fraud score?

In our check of 15 October 2026, yes on proxycheck.io: 50 of 55 QuanticData mobile exits in 11 countries scored 0 out of 100, the median was 0 in every country, and none was flagged as a VPN, Tor or hosting address. IP2Location.io flagged none of the 55 as a proxy. Carrier addresses are shared by many real phones, which keeps them low on most scales.

### What does a 0 fraud score proxy mean?

It means one scoring service, at the moment of the lookup, found no risk signal for that address: on proxycheck.io, no proxy, VPN, Tor, hosting or compromised detection and no recent attacks. It is a snapshot of one address on one service, not a permanent property, and other services can score the same address differently.

### How do I check the fraud score of my mobile proxy?

Read the exit address through the proxy (for example with ipinfo.io/json), then look that address up off the proxy on the scoring service you care about, and note the service, its settings and the time. With a sticky session the address holds for the session, so one lookup covers the job.

### Why do fraud score checkers give different results for the same IP?

Each checker has its own data and model: some weigh recent attacks, some the share of suspicious traffic seen from the whole carrier, some lists of known proxy endpoints. For mobile addresses behind carrier-grade NAT those signals differ a lot, so the same address can read 0 on one service and higher on another.

### How much do QuanticData mobile proxies cost?

$5 for 1 GB valid 90 days, down to $2.30 per GB at 500 GB, as one-off packs with no subscription. Country, carrier, region and city targeting and sticky sessions are included in the price.

## Carrier exits with a measured reputation

Measured on 15 October 2026: a median risk score of 0 in all 11 countries we checked, 50 of 55 exits at 0 on proxycheck.io. Mobile packs start at $5 for 1 GB, valid 90 days, no subscription.

[Start free — $2/month included](https://quanticdata.io/signup/?utm_source=blog&utm_medium=website&utm_campaign=blog-end&utm_content=mobile-proxies-fraud-score-tested)[Explore 4G/5G Mobile Proxies: City & ISP Targeting](https://quanticdata.io/mobile-proxies/)

## Related reading

[Proxies Geo-Targeted Proxies: 987 Words UK vs 149 DE The same URL, rendered through residential exits: 987 words from the United Kingdom, 149 from Germany, 2,031 from the United States. Nothing was translated; each exit landed on a different licensed product. On 18 of the 53 sites we measured, geo-targeted proxies change the page itself: the domain on AliExpress, the entity on OKX, the price line on Netflix, the number format on Facebook and LinkedIn, and whether the channel page exists at all on YouTube. Pin the exit per job or you are comparing two different pages. Read more](https://quanticdata.io/blog/same-url-different-country-different-page/) [Proxies Headless Browser Scraping: 53 Targets Measured Headless browser scraping pays on 15 of the 53 sites we measured between 26 and 28 September 2026, and costs words on 12. Every target was fetched through a residential exit as a plain HTTP client and again fully rendered; the table below has both word counts, the verdict and the one-line setting for all 53. Fifteen need the browser, four gain a little, fourteen gain nothing for up to 15 times the bytes, twelve lose words, and eight return no rows in either mode and want a collector. Read more](https://quanticdata.io/blog/52-targets-measured-browser-or-not/) [Proxies Web Scraping API vs Proxy: 25 eBay Rows, 19 s Web scraping API vs proxy is the wrong question on eight of the 53 sites we measured. On Amazon, eBay, Booking.com, Skyscanner, Airbnb, Expedia, Taobao and Spotify the page a proxy fetches holds between 0 and 72 words in either fetch mode. On 28 September 2026 the ebay_search collector returned 25 priced listings in 19.5 seconds and booking_stays returned 20 priced Rome properties in 5.1 seconds. The table, the cost per row and the setting for each site. Read more](https://quanticdata.io/blog/when-the-collector-wins/)

---

Source: https://quanticdata.io/blog/mobile-proxies-fraud-score-tested/ · Site index for AI: https://quanticdata.io/llms.txt