Cache for solutions

I find that the non-first time of the same request could respond faster than the first time.
It seems that GH is using a cache to save the request and its solution.
Therefore I would like to know if the cache was used and where it was implemented.
Otherwise, how can I explain the shorter computation time of the second call of the same request?
Thank you very much.

Can you explain what you did and how you came to the conclusion that the second call is faster?

JVM warmup, JIT kicking in,…

A lot of reasons which might explain what you’re observing but there is no such cache

I started GH routing engine on localhost or on my server, created a request on Postman.
Comparing Response’s time on Postman, I could state that the non-first call spent less time than the first call, e.g. 1200ms vs. 350ms (400ms, 370ms, etc.)