Surfshark VPN Test Results

Our latest Surfshark VPN test showed strong nearby performance in Sweden, where Surfshark retained 95.1% of our VPN-off download speed. UK performance remained good at 83.8% retention, while Germany, US East and Singapore showed progressively larger throughput reductions. Across all five tested VPN regions, our fixed-target stability probe recorded 0% packet loss.

← All VPN testing results

PPG LAB TESTED · PAID PLAN

We tested Surfshark using WireGuard across all five Private Proxy Guide core VPN regions. Surfshark performed especially well in Sweden, retaining 95.1% of our VPN-off download speed. UK performance remained strong at 83.8%, while Germany, US East and Singapore showed larger throughput reductions. The stability probe recorded 0% packet loss in every tested region.

Test date: August 24, 2026 Test ID: PPG-VPN-2026-08-BD02 Plan: Paid App version: 6.16.0 Protocol: WireGuard Runner: v0.4.5
95.1%
Nearby speed retained
Sweden median download
16.7 ms
Sweden latency
+7.6 ms vs baseline
0%
Stability packet loss
Across all five VPN regions
5/5
Exit-country checks passed
All PPG core regions

What our Surfshark test showed

Surfshark’s strongest result was Sweden, where median download speed reached 89.81 Mbps against our 94.45 Mbps VPN-off baseline. That corresponds to 95.09% speed retention.

The UK also performed well at 83.83% retention. Germany was noticeably weaker at 62.99%, while US East and Singapore retained only 23.30% and 11.51% of baseline download throughput respectively.

Despite those long-distance speed reductions, tunnel stability remained consistent. All five VPN locations recorded 0% packet loss, no detected latency spikes and low stability jitter.

Surfshark speed test results

Our VPN-off baseline was 94.45 Mbps download, 94.58 Mbps upload and 9.12 ms latency. We performed five VPN-off baseline runs and three speed-test runs per Surfshark location. Median values are used for the reported results.

Download speed retained

Sweden95.1%
 
United Kingdom83.8%
 
Germany63.0%
 
US East23.3%
 
Singapore11.5%
 

Speed retention compares the median Surfshark download result with the 94.45 Mbps VPN-off baseline.

Quick interpretation

  • Sweden: excellent, essentially near-baseline throughput.
  • UK: strong result with 83.8% retained.
  • Germany: a more substantial reduction than the other European routes.
  • US East: large long-distance throughput reduction.
  • Singapore: the lowest throughput in this Surfshark batch.
Location Download Upload Speed retained Regional ping Added latency Ookla jitter
Sweden 89.81 Mbps 90.38 Mbps 95.09% 16.68 ms +7.56 ms 0.266 ms
United Kingdom 79.18 Mbps 78.54 Mbps 83.83% 27.20 ms +18.08 ms 0.269 ms
Germany 59.49 Mbps 59.56 Mbps 62.99% 35.12 ms +26.00 ms 2.185 ms
US East 22.01 Mbps 21.77 Mbps 23.30% 94.37 ms +85.25 ms 0.260 ms
Singapore 10.87 Mbps 10.57 Mbps 11.51% 190.36 ms +181.24 ms 0.182 ms

Gaming latency and connection stability

For gaming-related testing, we separate regional latency from tunnel stability. Regional ping comes from the locked Ookla test endpoint for each location. A separate 50-packet fixed-target probe measures p95 latency, jitter, packet loss and latency spikes.

Important: these figures describe Surfshark’s VPN network path. They are not measurements against a specific game’s own servers unless that infrastructure is explicitly tested.
Location Regional ping Stability median p95 Stability jitter Packet loss Spike rate
Sweden 16.68 ms 17 ms 18 ms 0.449 ms 0% 0%
United Kingdom 27.20 ms 27 ms 27 ms 0.000 ms 0% 0%
Germany 35.12 ms 35 ms 35 ms 0.469 ms 0% 0%
US East 94.37 ms 94 ms 94 ms 0.429 ms 0% 0%
Singapore 190.36 ms 190 ms 190 ms 0.224 ms 0% 0%

Surfshark’s stability results were notably tight even on the long-distance routes. US East had a 94 ms median and 94 ms p95, while Singapore had a 190 ms median and 190 ms p95. No packet loss or latency spikes were detected in any tested region.

Connection time and location validation

The runner measures connection time by detecting the transition to the VPN public IP after the connection process begins. Every Surfshark test region returned the expected exit country.

Location Detected connect time Expected country Location validation
Sweden 3.17 s SE PASS
United Kingdom 4.28 s GB PASS
Germany 3.55 s DE PASS
US East 5.19 s US PASS
Singapore 6.13 s SG PASS

Detected connection timing reflects the public-IP transition observed by the test runner and should not be interpreted as a universal connection time for every Surfshark device or network.

DNS leak testing

IPLeak.net DNS testing was enabled for this Surfshark batch. The runner first recorded the resolver visible with the VPN disconnected and then compared it with resolver IPs observed while connected to each Surfshark location.

Location DNS result Interpretation
Sweden PASS No resolver IP seen during the VPN-off baseline was reused.
United Kingdom PASS No resolver IP seen during the VPN-off baseline was reused.
Germany PASS No resolver IP seen during the VPN-off baseline was reused.
US East PASS No resolver IP seen during the VPN-off baseline was reused.
Singapore PASS No resolver IP seen during the VPN-off baseline was reused.

All five regions returned the automated classification PASS_NO_BASELINE_RESOLVER_REUSED. This test checks whether the VPN-off resolver is still visible; it does not by itself establish ownership of every DNS resolver involved.

IPv4, IPv6 and security observations

  • IPv4 exit validation: PASS in all five Surfshark locations. The public IPv4 changed and the detected country matched the requested PPG region.
  • IPv6: INCONCLUSIVE BASELINE. The baseline IPv6 endpoints could not be resolved, so this test batch cannot reliably establish baseline IPv6 availability or make a Pass/Fail claim about IPv6 handling.
  • DNS resolver comparison: PASS across all five tested locations.
  • BrowserLeaks: not enabled in this Surfshark batch.
  • Kill switch: not tested in this batch.
Why IPv6 remains inconclusive: the IPv6 lookup endpoints failed because their hostnames could not be resolved. We do not treat a failed baseline endpoint as evidence that IPv6 was unavailable or protected by the VPN.

Test evidence

The Surfshark test produced raw speed measurements, fixed-target stability data, DNS resolver evidence and automated screenshots for the baseline and all five VPN locations.

26/26 Automated desktop and Speedtest screenshots completed.
20 Ookla runs: 5 VPN-off baseline runs and 15 Surfshark runs.
6 IPLeak DNS evidence sets: baseline plus five VPN regions.

Raw screenshots and evidence may contain public IP addresses or environment-specific information and should be reviewed before public publication.

How we tested Surfshark

  • Plan: Surfshark paid plan.
  • Application: Surfshark 6.16.0 for Windows.
  • Protocol: WireGuard.
  • Baseline: 5 VPN-off Ookla runs.
  • Core PPG regions: Sweden, United Kingdom, Germany, US East and Singapore.
  • Core-region coverage: 5/5.
  • Speed testing: 3 runs per VPN region, reported using medians.
  • Regional Speedtest locking: enabled.
  • Post-connect stabilization: 20 seconds.
  • Gaming stability: 50 ICMP samples against the fixed stability target.
  • IPLeak.net DNS test: enabled.
  • Automatic screenshots: enabled.
  • BrowserLeaks evidence: not enabled.
  • Kill-switch test: not enabled.

Raw Surfshark speed-test runs

The summary tables above use median values. The individual measurements used to calculate those medians are available below.

Show all 20 individual speed-test runs

Baseline – VPN disconnected

Run Download Upload Ping Jitter Loss Ookla server
#1 93.17 Mbps 94.83 Mbps 17.96 ms 7.353 ms 0% Telenor Denmark, Søborg
#2 94.86 Mbps 94.73 Mbps 9.07 ms 0.273 ms 0% Telenor Denmark, Søborg
#3 94.73 Mbps 93.98 Mbps 9.14 ms 0.362 ms 0% Telenor Denmark, Søborg
#4 94.22 Mbps 94.58 Mbps 9.12 ms 0.139 ms 0% Telenor Denmark, Søborg
#5 94.45 Mbps 94.46 Mbps 9.09 ms 0.093 ms 0% Telenor Denmark, Søborg

Sweden

Run Download Upload Ping Jitter Loss Ookla server
#1 89.51 Mbps 90.25 Mbps 16.68 ms 0.266 ms 0% Tele2 Sweden, Stockholm
#2 89.81 Mbps 90.38 Mbps 16.62 ms 0.209 ms 0% Tele2 Sweden, Stockholm
#3 89.85 Mbps 90.49 Mbps 16.77 ms 0.682 ms 0% Tele2 Sweden, Stockholm

United Kingdom

Run Download Upload Ping Jitter Loss Ookla server
#1 78.67 Mbps 77.69 Mbps 48.88 ms 10.210 ms 0% Structured Communications, London
#2 79.18 Mbps 79.18 Mbps 27.11 ms 0.101 ms 0% Structured Communications, London
#3 79.84 Mbps 78.54 Mbps 27.20 ms 0.269 ms 0% Structured Communications, London

Germany

Run Download Upload Ping Jitter Loss Ookla server
#1 59.32 Mbps 56.80 Mbps 35.57 ms 2.185 ms 0% 23M GmbH, Frankfurt am Main
#2 60.26 Mbps 59.56 Mbps 35.12 ms 0.083 ms 0% 23M GmbH, Frankfurt am Main
#3 59.49 Mbps 59.72 Mbps 34.97 ms 3.495 ms 0% 23M GmbH, Frankfurt am Main

US East

Run Download Upload Ping Jitter Loss Ookla server
#1 22.01 Mbps 21.77 Mbps 94.37 ms 0.230 ms 0% Natural Wireless, New York, NY
#2 22.01 Mbps 21.80 Mbps 94.62 ms 0.260 ms 0% Natural Wireless, New York, NY
#3 22.02 Mbps 21.72 Mbps 94.27 ms 0.316 ms 0% Natural Wireless, New York, NY

Singapore

Run Download Upload Ping Jitter Loss Ookla server
#1 10.86 Mbps 10.57 Mbps 190.36 ms 0.174 ms 0% Symphony Communication PCL, Singapore
#2 10.88 Mbps 10.93 Mbps 189.95 ms 0.182 ms 0% Symphony Communication PCL, Singapore
#3 10.87 Mbps 8.19 Mbps 191.24 ms 0.554 ms 0% Symphony Communication PCL, Singapore

Surfshark test conclusion

Surfshark performed very well on our closest test route. Sweden retained 95.09% of baseline download speed, with 16.68 ms regional latency, an 18 ms stability p95 and 0% packet loss.

The UK also produced a strong result at 83.83% speed retention. Germany was less efficient in this batch at 62.99%, while the long-distance US East and Singapore routes retained 23.30% and 11.51% respectively.

One of Surfshark’s strongest characteristics in this test was consistency. All five core regions recorded 0% packet loss, no detected stability spikes and successful DNS resolver separation from the VPN-off baseline.

This dataset describes Surfshark’s standard paid Windows connection using WireGuard. It does not establish performance for MultiHop, Static IP, Dedicated IP, streaming unblocking, torrenting, specific game servers or kill-switch behavior because those were not separately tested in this batch.

← Back to all VPN testing results