That was a ballpark estimation, and while pure req/sec is usually irrelevant, it still comes in when the load starts to increase and you need to squeeze every last drop of perf.

PHP usually is doing poorly, even with its "cache", and when you realize that PHP devs actually are not PHP devs, but framework (like laravel/symfony) devs you need to account for the overhead of the framework too.

But you can have a look at the framework benchmarks game:

https://www.techempower.com/benchmarks/#section=data-r21&hw=...