If you mean processing more requests/second than v1, then write a program that makes random (but valid, or perhaps a combination of valid and invalid) requests to v1 and then run it against both v1 and v2 to see what the minimum time is.
If by “better,” you mean faster responses, then perhaps do the same thing at the same rate and calculate the differences in time between the submission of a request to receiving its response and find out which API has the lowest difference between those times.