sqlite3 persistent connection vs single cli call and how to implement former

Question

This question is in part due to my lack of knowledge of the proper terminology but I can explain this practically:

It is possible to simply call sqlite3 mydatabase, and then from the command promt enter

> insert into people(firstname) values ('Joe');
> insert into people(firstname) values ('Bob');
... etc - let's say 1000 entries

It is also possible to run the following command line commands:

sqlite3 /path/to/mydatabase "insert into people(firstname) values('Joe');" ".exit"
sqlite3 /path/to/mydatabase "insert into people(firstname) values('Bob');" ".exit"

And perhaps even the last ".exit" is not needed (I am not sure).

From a performance perspective (if I could type infinitely fast) it seems the former would be faster than calling the 1000 commands in the second example.

My question is, is there a way to do the first example via a persistent object, ie be as if I was typing an unending series of queries at the sqlite prompt without exiting? I'm open to doing it in php, python or go (in that order).