Precise benchmarking is a complex topic. The following results are intended more as an illustration of the difference in performance between platforms and configurations than an in-depth exploration of performance.

If performance is important to you then we highly recommend running your own profiling.

These tests are run using a console performance example. Evidence from the Evidence Records file in device-detection-data is processed by the pipeline and the time taken measured. If language appropriate, multiple threads are used to run the tests. For technical details, see the example repositories on GitHub. We run performance tests regularly on GitHub actions. The performance of this may not be as good as a dedicated machine, as it's running on a GitHub runner, but gives an indication of consistency.

C/C++

.NET

Node.js

Java

Python