Floppydata Floppydata

Mobile Proxies with the Least Shared IPs: Pool Overlap Tested

Only 5.5% of Floppydata's 23,743 exit IPs were also given out by another provider in our mobile benchmark, the lowest of the 11 providers. IPRoyal had 10.1%; the others 26.9%–56.4%.

We operate Floppydata and tested it with the same scripts, servers and schedule as the other 10 providers. The full data and scripts are public.

See the ranking
11 providers
28,800 requests each
12 exit countries
24–27 Oct 2026
Mobile Benchmark

The ranking

Mobile Proxies with the Least Shared IPs: Pool Overlap Tested: all 11 providers
# Provider IPs shared with other providers lower is better Unique exit IPs whole test
1 Floppydata 5.5% 95% CI 5.2–5.8 23,743
2 IPRoyal 10.1% 95% CI 9.7–10.5 24,661
3 ResiProx 26.9% 95% CI 26.3–27.6 19,215
4 Decodo 31.5% 95% CI 30.9–32.1 23,323
5 Oxylabs 31.7% 95% CI 31.1–32.3 23,216
6 SOAX 36.8% 95% CI 36.1–37.6 16,450
7 DataImpulse 41.5% 95% CI 40.7–42.3 14,507
8 Evomi 52.8% 95% CI 52.1–53.5 18,270
9 ProxyEmpire 53.3% 95% CI 52.7–54.0 22,848
10 Byteful 55.7% 95% CI 55.0–56.4 19,989
11 NodeMaven 56.4% 95% CI 55.7–57.1 21,830

First place Floppydata (we operate it) = tied place

The share of a provider's unique exit IPs (successful rotating requests over the whole test) that at least one of the other 10 tested providers also gave out. IPs are compared as anonymized hashes; the raw addresses are not published.

Check the numbers yourself: the anonymized raw data, scripts and full results workbook are on GitHub.

Key findings

Some pools overlap a lot: up to a third of the IPs of one provider came out of another provider too.

  • Floppydata has the least shared pool

    5.5% of its 23,743 exit IPs were seen at another provider, then IPRoyal at 10.1%.

  • Byteful and NodeMaven overlap most

    6,572 shared IPs, 30–33% of their two pools; 72.6% of them were given out by both within the same hour.

  • Decodo and Oxylabs

    5,866 shared IPs, 25–25% of their two pools.

  • Not just the carrier

    A shared IP can be carrier-grade NAT, where many phones use one address. But providers that exit through the same carriers in the same countries overlap very differently, from 5.5% to 56.4%, so NAT alone does not explain it. We do not claim that any two providers share a pool.

How we measured

How we measured shared IPs

Every provider got 28,800 rotating requests to a Cloudflare trace endpoint on 24–27 October 2026: 100 per exit country, 12 countries, 3 servers, 8 runs. Cloudflare returns the exit IP of each request.

  • Shared IP. An exit IP of one provider that any other tested provider also returned at least once during the test.
  • The share. Shared IPs ÷ unique exit IPs of the provider, with a Wilson 95% confidence interval. Providers whose interval overlaps the leader's share first place.
  • Same hour. For the pairs that overlap most, we also counted how many shared IPs both providers gave out within the same hour.

Limits

  • Only these 11 providers. The share counts overlap with the providers in this test. Overlap with untested providers is unknown.
  • Carrier NAT. Mobile carriers put many phones behind one public IP, so some overlap is normal and does not mean two providers share devices.
  • Raw data and scripts. The anonymized request logs, the test scripts and the results workbook are on GitHub, so every number here can be recomputed.

More from the benchmark

Download raw benchmark data

Anonymized request logs, the test scripts and the full results workbook. Check every number on this page yourself.

Frequently Asked Questions

Do mobile proxy providers share the same IPs?
Partly. In our benchmark 5.5% of Floppydata's exit IPs were also given out by another provider, and up to 56.4% at NodeMaven. Byteful and NodeMaven shared 30–33% of their pools.
Why does a shared IP matter?
An IP that several pools hand out carries more strangers' traffic, so its reputation on a site depends on more people than one provider's customers.
Is a shared IP a sign of reselling?
Not necessarily. Many phones can sit behind one carrier IP. Our data shows the overlap, not its cause, so we do not draw conclusions about any provider.