Yep, I did a test tonight and compare only Python and Go. This time Go is “only” 2 times faster. That’s a great improvement for Python. But it’s especially a performance loss for Golang that must used math/big library to manage bigint.

That’s not enough relevant because we are going in the Go worst case (and I don’t want to depend on other libraries in the algorithm to have something comparable)

It’s definitively difficult to find a fair test without reaching the edge cases of each language.