Who actually operates the residential proxy networks
Founding year, headquarters and claimed pool size across 27 providers. Snapshot 13 August 2026.
The table
| Provider | Founded | Headquarters | Claimed pool | Countries | Buyable |
|---|
| Oxylabs | 2015 | Lithuania | 175M | 195 | Active |
| SOAX | 2019 | United Kingdom | 155M | 190 | Active |
| Bright Data | 2014 | Israel | 150M | 195 | Active |
| ProxyScrape | 2018 | Netherlands | 120M | 130 | Active |
| Decodo | 2018 | Lithuania | 115M | 195 | Active |
| IPBurger | not published | not published | 100M | 195 | Active |
| Thordata | not published | not published | 100M | 190 | Active |
| DataImpulse | 2022 | Ukraine | 90M | 194 | Active |
| Proxy-Cheap | not published | not published | 80M | 100 | Active |
| Evomi | not published | CH | 54M | not published | Active |
| ProxyRack | not published | not published | 35M | not published | Active |
| IPRoyal | 2020 | Lithuania | 34M | 195 | Active |
| Shifter | 2012 | RO | 31M | 130 | Active |
| Webshare | 2018 | United States | 30M | 195 | Active |
| NodeMaven | 2022 | Estonia | 30M | 150 | Active |
| Rayobyte | 2015 | United States | 25M | 150 | Active |
| 9Proxy | not published | not published | 20M | not published | Paused |
| Proxy-Seller | 2016 | Cyprus | 18M | 100 | Active |
| Infatica | 2019 | Singapore | 15M | 150 | Active |
| Froxy | 2021 | Cyprus | 10M | 100 | Active |
| ProxyEmpire | 2020 | United States | 9M | 170 | Active |
| PacketStream | 2018 | United States | 7M | 190 | Active |
| ASocks | not published | not published | 7M | 150 | Active |
| Geonode | 2020 | United States | 2M | 140 | Active |
| Storm Proxies | 2016 | United States | 0.2M | not published | Active |
| PrivateProxy | 2011 | United Kingdom | 0.1M | 30 | Active |
| Massive | not published | not published | not published | 195 | Active |
Read the pool figures as claims, not measurements
Every number in the pool column is what the provider says, and the column is labelled claimed for that reason. Nobody outside these companies can audit an IP pool, the figures are marketing inputs, and they are frequently rounded up and rarely dated. Added together the claims come to roughly 1412 million addresses. Overlap between providers is certain, because the same consumer devices are resold through multiple networks, so that total should not be read as distinct IPs.
The useful question is not how big a pool is. It is where the addresses came from, and 13 of the 27 publish nothing at all on that.
What is missing, and why it is left blank
Founding year is on record for 19 of 27, headquarters for 20, pool claims for 26. The rest publish nothing legible and are marked not published rather than filled with a guess. A corporate identity that is hard to establish is itself a finding, particularly in a market where one operator was seized by the FBI in July 2026 and its network had already been resold under other brands.
Related