#include <iostream>
#include <vector>
#include <set>
#include <queue>
#include <stack>
#include <math.h>
#include <cassert>
#include <bitset>
#include <map>
#include <algorithm>
#include <iterator>
#include <string>
#include <utility>
#include <numeric>
#include <regex>
#include <complex>
#include <random>
#include <atcoder/all>
using namespace atcoder;
using namespace std;
using ll=long long;
using ull=unsigned long long;
using P=pair<ll,ll>;
template<typename T>using minque=priority_queue<T,vector<T>,greater<T>>;
template<typename T>bool chmax(T &a,const T &b){return (a<b?(a=b,true):false);}
template<typename T>bool chmin(T &a,const T &b){return (a>b?(a=b,true):false);}
template<typename T1,typename T2>
ostream &operator<<(ostream &os,const pair<T1,T2>&p){
  os<<p.first<<" "<<p.second;
  return os;
}
template<typename T1,typename T2>
istream &operator>>(istream &is,pair<T1,T2>&p){
  is>>p.first>>p.second;
  return is;
}
template<typename T>
istream &operator>>(istream &is,vector<T> &a){
  for(auto &i:a)is>>i;
  return is;
}
#define reps(i,a,n) for(int i=(a);i<(n);i++)
#define rep(i,n) reps(i,0,n)
#define all(x) x.begin(),x.end()
ll myceil(ll a,ll b){return (a+b-1)/b;}
using mint=modint1000000007;
int main(){
  int n,p;
  cin>>n>>p;
  mint ans=0;
  ll cu=p;
  while(true){
    ans+=n/cu;
    cu*=p;
    if(n<cu)break;
  }
  mint cnt=1;
  reps(i,1,n+1)cnt*=i;
  reps(i,1,n+1){
    cnt=cnt.pow(i);
  }
  cout<<(ans*cnt).val()<<endl;
}