Review

Oxylabs Review

A large, very reliable network at a price that only works above a terabyte a month.

Oxylabs
8.3 /10
Live pool
9
Network spread
9.2
Price per GB
5.9
Response time
8.8
Country coverage
10
Features
9.5
Transparency
7

The short version

  • Oxylabs returned 140,811 live US addresses in a single window and led three of the five countries we measured, but the margins were narrow: 3% in the United States, under 10% in France and Germany.
  • Responses were consistent but not fast, 646 ms to 752 ms at the median, roughly twice our own timings on the same targets from the same machine.
  • Network spread is where the pool thins out. In France its addresses came from 93 networks against our 148, so a large pool sits on comparatively few operators.
  • Price is the deciding factor. The $2.50 per GB headline needs a $2,500 monthly commitment, and the entry tier is $6 per GB.

At a glance

Founded 2015
Headquarters Vilnius, Lithuania
Advertised pool 175M+ residential IPs
Countries 195+
Live IPs measured 276,680 across five countries
Protocols HTTP(S), HTTP3, SOCKS5
Authentication Username and password, or IP whitelist
Sessions Rotating, or sticky up to 24 hours
Concurrency Unlimited concurrent sessions
Entry price $30 for 5 GB ($6.00/GB)
Best published rate $2.50/GB at 1 TB
Free trial On request, once per customer

What we measured

Live IPs each network had answering during the same test window, five countries, identical request volumes and concurrency for both sides.

Country Oxylabs Shifter Networks
United States 140,811 136,667 1,591
United Kingdom 46,394 53,506 188
France 36,888 33,581 93
Canada 28,783 33,485 228
Germany 23,804 22,005 183
See the full benchmark and method →

How big is the pool, really

Oxylabs advertises 175M+ residential IPs. Like every advertised pool figure in this industry, that is a lifetime total: every address the network has ever seen, across every country, accumulated over years. It is a real number and a useful measure of reach, but it is not the number that decides whether your crawl finishes. So we measured that one instead: how many distinct addresses actually answer in a given country during a single test window.

Across five countries Oxylabs served 276,680 live IPs. In the United States it returned 140,811 from 250,000 requests, the highest single-country figure in the series so far, though only 3% clear of the 136,667 we recorded ourselves in the same window. It also led France with 36,888 against 33,581, and Germany with 23,804 against 22,005, gaps of under 10%.

The United Kingdom and Canada went the other way. There Oxylabs returned 46,394 and 28,783 against our 53,506 and 33,485. So the shape of the network is uneven, and no single country was decided by much. Across all five, Oxylabs served 276,680 live addresses against our 279,244, a difference of under 1%.

Advertised totals count every address a network has ever seen. Residential IPs rotate all day as real devices come and go, so both figures are real, they just answer different questions. Ours answers what you can reach at once.

Reliability and speed

This is the part of the review where Oxylabs is simply the best network we have tested. Nothing else in the series came back that clean.

If your pipeline is unattended and a failed request costs you a retry, a queue slot or a partial dataset, that consistency is worth real money and it is the strongest argument for paying their rate.

Speed is a different story. Median response times ran from 646 ms in Germany to 752 ms in the United States. Consistent, but roughly twice our own medians of 324 ms to 445 ms on the same targets from the same machine. For interactive work or anything latency-bound, that gap compounds.

Network spread, and why it matters more than pool size

A pool of a given size behaves very differently depending on how many separate network operators it sits across. Anti-bot systems key on the network an address belongs to, so a hundred thousand addresses spread over 1,600 operators is harder to fingerprint than the same count concentrated on 200.

Oxylabs is strong here in the United States, 1,591 networks against our 1,606, effectively level. Europe is where it thins: 93 networks in France against our 148, and 183 in Germany against our 276. Canada is the one country where it leads us outright on spread, 228 against 218.

Read alongside the pool figures, that produces the honest summary of this network: in France it holds more addresses than we do across barely two thirds of the operators.

What it actually costs

The rate quoted in most comparisons is $2.50 per GB. That is real, and it requires the Corporate plan at $2,500 a month for a terabyte. Below that the ladder is $4 per GB at 125 GB, $5 at 20 GB and $6 at the 5 GB entry tier.

So the honest way to read Oxylabs pricing is by the volume you will actually buy. A team using 100 GB a month is looking at roughly $4 per GB, not $2.50, which puts it among the most expensive options in the market at that size.

For context on the same page of the market, our own rate is $2.00 per GB at entry and $0.75 at volume, and we publish the whole ladder without a sales call. That is a disclosed interest, not a neutral observation, which is exactly why the performance figures above come from a test anyone can repeat.

Buying and running it

Oxylabs runs KYC on residential proxy accounts and enforces a prohibited-domain list, which is normal for the enterprise tier of this market and slower than self-serve rivals. The free trial exists but is granted on request rather than by signing up.

The dashboard covers sub-users, access control and usage reporting, and the documentation is among the better sets in the industry. Integration is standard: a gateway host, credentials, and country or city targeting through parameters, so moving to or from it is a configuration change rather than a rewrite.

Pricing

Plan Traffic Per GB Monthly
Starter 5 GB $6.00 $30
Basic 20 GB $5.00 $100
Advanced 125 GB $4.00 $500
Corporate 1 TB $2.50 $2,500

Pros and cons

Pros
  • A very large US pool, 140,811 live addresses, narrowly the highest we have measured
  • Near-total country coverage at 195+, with city and ASN targeting
  • Near-total country coverage at 195+
  • Unlimited concurrency, sticky sessions up to 24 hours, HTTP3 and SOCKS5
  • Strong documentation, sub-user controls and enterprise support
Cons
  • $2.50 per GB needs a $2,500 monthly commitment; real-world rates are $4 to $6
  • Median responses roughly twice ours on identical targets
  • Thin network spread in Europe, 93 operators in France against our 148
  • KYC and a request-only trial slow down evaluation
  • Pool leadership is narrow and uneven: ahead in three of five countries by under 10%, behind in two

Who Oxylabs is for

Buy it if
  • You run enterprise volume, a terabyte a month or more, where the $2.50 rate is actually reachable.
  • You need the deepest possible US pool in a single window and are not price sensitive.
  • You want depth in the United States above all else, and are not price sensitive.
  • You want compliance paperwork, a dedicated account manager and formal SLAs.
Look elsewhere if
  • You are under a terabyte a month, where the effective rate is $4 to $6 per GB.
  • Latency matters to your job, since medians ran roughly twice ours.
  • You need broad network diversity in Europe rather than raw address count.
  • You want to start small without talking to sales.

How we scored it

Criterion Weight Score Basis
Live pool measured 25% 9 276,680 IPs live across five countries, against the tested field.
Network spread measured 20% 9.2 2,283 distinct networks across five countries.
Price per GB measured 25% 5.9 $2.50 per GB, which requires the 1 TB Corporate plan at $2,500 a month. A 125 GB buyer pays $4.00 and the entry tier is $6.00.
Response time measured 10% 8.8 700 ms median across the countries tested.
Country coverage 10% 10 195+ countries advertised, scaled against full global coverage.
Features 5% 9.5 HTTP(S), HTTP3 and SOCKS5, unlimited concurrency, 24 hour sticky sessions, city and ASN targeting, sub-user access control. Close to the most complete feature set in the market.
Transparency 5% 7 The full price ladder and pool claims are published without a sales call, which many rivals do not do. Marked down because the trial is request-only and the headline rate needs a commitment that is easy to miss.
FAQ

Frequently asked questions

Common questions about Oxylabs.

Below that you pay $4 to $6 per GB, and at that level the pool advantage does not survive the price difference against cheaper networks.

They advertise 175M+ residential IPs. In our own test, 250,000 requests through their US gateway returned 140,811 distinct live addresses, and five countries together returned 276,680. Advertised totals count every address ever seen; ours counts what answered during the test window.

It is consistent rather than fast. Medians ran 646 ms to 752 ms across the five countries, with almost no failed requests. On the same targets from the same machine our own medians were 324 ms to 445 ms.

Yes, once per customer, granted on request through their contact form or support rather than automatically at signup. Residential accounts also go through KYC.

Decodo measured within 2% of Oxylabs on live pool at $2.00 per GB. Webshare returned a larger five-country pool at $1.40. Our own network, Shifter, measured 279,244 live IPs across the same five countries at $0.75 per GB, and led the United Kingdom and Canada while trailing in the United States, France and Germany.

Only as far as it is checkable, which is why the method is published and the score is calculated rather than awarded. Six of the eight criteria come from measurements normalised across every network we have tested, so Oxylabs scores above us on pool, reliability and coverage. The full run is at /benchmark/oxylabs.

Get started

Test both networks on your own targets

Our numbers are a starting point, not a verdict. Measure what each network returns on the sites you actually scrape.

Try Shifter for FreeSet up in minutes. Cancel anytime.