NordLynx speed test: what it is and how it performs
Observed NordLynx performance across the Secure-VPN benchmark network. The page emphasizes normalized speed retention so different source-line speeds do not automatically dominate the comparison.
NordLynx by provider
| Provider | Samples | Countries | Median retention | Median download | Median latency |
|---|---|---|---|---|---|
| NordVPN | 4,932 | 65 | 61.9% | 54.1 Mbps | 122.2 ms |
Historical views
What is NordLynx?
NordLynx is NordVPN's WireGuard-based VPN protocol. It uses WireGuard as the underlying tunneling technology but adds NordVPN's own connection and privacy architecture around it, so Secure-VPN labels the measured protocol as NordLynx rather than treating it as identical to stock WireGuard.
What does a NordLynx tunnel mean?
When the NordVPN app says it is using a NordLynx tunnel, the encrypted VPN connection is using NordLynx for transport between your device and the VPN server. The route, exit location and source network still matter, so the protocol name alone does not guarantee a particular speed.
NordLynx vs OpenVPN
The live data should decide the performance question, not a protocol label. Open Speed by protocol to compare NordLynx with NordVPN's OpenVPN UDP and TCP results from one test location at a time. For the transport-level concepts, read WireGuard vs OpenVPN and OpenVPN UDP vs TCP.
NordLynx and OpenVPN should be compared with the same source worker and time window. Mixing a NordLynx result from one country with an OpenVPN result from another can turn route quality into a fake protocol advantage. The live protocol table keeps that source context visible.
NordLynx speed by destination
A protocol-wide median is only the starting point. Long-distance routes can add latency or reduce retained throughput even when the protocol performs well locally. Use the linked NordVPN provider profile and destination benchmark pages to see where the samples came from before applying the result to your own route.
How to read the NordLynx speed result
Median retention is the middle normalized result, while P10 and P90 show the slower and faster ends of the central distribution. Raw download speed is still useful, but retention is better for comparing test sources with different baseline connections.
If P10 is far below the median, some valid runs were substantially slower than the typical result. If P90 is near 100%, the faster end of the distribution retained almost all of its paired baseline. Neither value should be read without the sample count and destination coverage shown above.
