Is there a way to determine the objective latency of an API call?

Question

Context

I am currently researching latency with regards to Deep learning algorithms. My API call which invokes a post request to the classifier.

<