Why are C++ and Python code comparably slow? How to fix?

Question

Python

def main():
    n = int(input())
    input1 = list(input().split())
    output = []
    input1.reverse()
    for i in input1:
        if i not in output:
            output.append(i)

    print(len(output))
    output.reverse()
    for i in output:
        print(i, end = ' ')

main()

Similar code in C++

#include <iostream>
#include <vector>
#include <algorithm>
using namespace std;

bool checkel(vector<int> mas,int el)
{
    for (int i:mas)
        if (i == el)
            return true;
    return false;
}

int main()
{
    int n;
    cin >> n;
    int* mas = new int[n];
    for (int i=0;i<n ;i++ ) {
        cin >> mas[i];
    }
    vector<int> vec;
    for (int i=n-1;i>=0 ;i--) {
        if (!checkel(vec,mas[i])) {
            vec.push_back(mas[i]);
        }
    }

    cout << vec.size() << endl;
    reverse( vec.begin(), vec.end() );
    for (int i:vec)
        cout << i << ' ';
    return 0;
}

They showed commensurately poor results for speed. From Python I expected, but from crosses ...
Or maybe it's because of the output delay (in crosses)? How can it be optimized?

Answer 1

Perhaps it's a non-optimal algorithm. In the worst case, you will have quadratic complexity. Stop walking on the internal list all the time, and I think it will become faster. Well, you don't need to reverse the arrays, just iterate over them in reverse order.
Also look at how olympiads set up input-output, it can influence.

Answer 2

I don't think python is slow

Here you go.....

Answer 3

C++ code:
1. pass a vector to a function by reference, try a reference in the for loop - although int is unlikely to be profitable there
2. if you just need to output the result, then reverse is fashionable