Shifter Now Has the Largest Live US Residential IP Pool

On 7 October 2026 we ran Shifter back through our standard residential proxy benchmark. In 250,000 requests to an IP echo endpoint, Shifter’s US gateway returned 212,212 distinct live IPs. That is the largest live US pool of all 16 networks we have benchmarked, ahead of Webshare at 150,512, Oxylabs at 140,811, Proxy-Seller at 139,933, Infatica at 139,334 and Decodo at 138,914.

It is 41.0% more live US addresses than the next network on the list, and it comes at a lower price. Shifter starts from $0.10 per GB, against $1.40 for Webshare, $2.00 for Decodo, $2.20 for Proxy-Seller, $2.50 for Oxylabs and $2.60 for Infatica at their lowest published rates.

The US result also came with the widest network spread we recorded: 1,828 distinct ASNs, against 1,591 for Oxylabs, the next highest.

The US ranking

Every network below received the same 250,000 US requests. The table is sorted by live IPs, the number of distinct exit addresses each gateway handed back.

Infatica's gateway accepts only HTTPS, so its counts come from a run over a CONNECT tunnel and its latency is left out: each of its requests carries a TLS handshake the others do not.

Shifter leads on live IPs and on ASNs. Oxylabs, SOAX, Decodo and Webshare reached marginally more US cities, by 4.2% or less, which matters far less than a live pool 41% to 89% larger than theirs. Rayobyte answered a little faster on average, at 333 ms against 427 ms.

Bright Data is not in the benchmark. It has no self-serve residential access, so there are no measured figures for it.

What “live IPs” means, and why it matters for crawling

Advertised pool sizes count every address a network sees over a long window. Shifter’s own figure is more than 205 million addresses over a month. That number says little about the crawl you are running right now, because residential devices come and go through the day.

A live IP count answers a narrower question: how many distinct addresses did the gateway actually hand out, in one country, during one test window? Each fresh connection asks the gateway for a new exit, so every distinct address that comes back was live and reachable at that moment. The result is a lower bound on the pool that was available, not a census of the network.

For crawling, that is the number that sets your limits. The more distinct live addresses you can draw from in a country, the fewer requests land on each one, and the less any single target sees from a single IP. Spread matters too: addresses drawn from 1,828 different networks are harder to block as a group than the same number concentrated in a few large ISPs.

The other four countries

We ran the same test in the United Kingdom, France, Canada and Germany. Webshare was the strongest competitor in all four, so the table compares Shifter with Webshare.

Shifter ranked first on live IPs in the United Kingdom, France and Canada. It also reached the most distinct networks in France (147 ASNs) and Canada (308 ASNs), while Webshare reached more in the United Kingdom, 209 against 166. Shifter’s average latency was 447 ms in the United Kingdom, 443 ms in France and 377 ms in Canada, against 875 ms, 927 ms and 937 ms for Webshare.

Germany is the exception. Shifter placed fourth there with 22,428 live IPs, behind Webshare (25,202), Oxylabs (23,804) and Decodo (23,580). The gap to third place was 1,152 addresses, about 5%. Shifter’s German IPs did come from 272 networks, level with Rayobyte for the widest spread in that country.

Across all five countries Shifter returned 390,093 live IPs, against 306,410 for Webshare, the next highest total.

What changed since August

In our August run, Shifter returned 136,667 live US IPs. On 7 October it returned 212,212, an increase of 75,545 addresses, or 55%. The other countries moved up as well: the United Kingdom from 53,506 to 60,487, France from 33,581 to 49,449, Canada from 33,485 to 45,517 and Germany from 22,005 to 22,428.

That is the measurement. Both runs used the same method, the same server and the same volumes, so the figures are directly comparable.

Method and dates

The method is identical for every network. Each request was plain HTTP to an IP echo endpoint, over a fresh connection through the provider’s rotating residential gateway. We used 1,000 concurrent connections, a 10 second timeout and fixed volumes per country: 250,000 requests for the United States, 100,000 for the United Kingdom, 80,000 for France, 70,000 for Canada and 60,000 for Germany. Providers ran one at a time, from the same server, so no two competed for bandwidth.

The competitor figures come from their own runs between 5 August and 4 September 2026, and each network’s benchmark page shows its test date. Shifter was re-measured on 7 October 2026 with the identical method, server and settings. Residential pools change hour by hour, so treat every figure as a snapshot of what was live at the time.

The full method, including the failures, is in how we benchmarked every major residential proxy network.

How to reproduce it

You do not need to trust our numbers. Point any HTTP client at an IP echo endpoint through a provider’s rotating gateway, open a new connection for every request so the gateway rotates, and count the distinct addresses that come back:

export PROXY_URL="http://USER:PASS@GATEWAY:PORT"
seq 1 10000 | xargs -P 100 -n1 sh -c \
  'curl -s --max-time 10 -x "$PROXY_URL" http://ip-info.com/json | jq -r .ip' \
  | sort -u | wc -l

Use the same request count, the same machine and the same hour for every provider you compare, and test in the countries you actually crawl.

See the full results

Every network has its own page with the country by country breakdown, including Shifter vs Webshare, Shifter vs Oxylabs and Shifter vs Decodo. The complete set is on the benchmarks page, and plans start at $0.10 per GB on the Residential Proxies pricing page.