//GIVE ME AC!!!!!!!!!!!!!!!!!
//#pragma GCC target("avx")
//#pragma GCC optimize("O3")
//#pragma GCC optimize("unroll-loops")
#include<bits/stdc++.h>
#define ll long long
#define ld long double
#define floatset() fixed<<setprecision(15)
#define all(n) n.begin(),n.end()
#define rall(n) n.rbegin(),n.rend()
#define rep(i, s, n) for (ll i = s; i < (ll)(n); i++)
#define pb push_back
#define eb emplace_back
#define max_(a) *max_element(all(a))
#define min_(a) *min_element(all(a))
#define INT(...) int __VA_ARGS__;scan(__VA_ARGS__)
#define LL(...) ll __VA_ARGS__;scan(__VA_ARGS__)
#define STR(...) string __VA_ARGS__;scan(__VA_ARGS__)
#define CHR(...) char __VA_ARGS__;scan(__VA_ARGS__)
#define DBL(...) double __VA_ARGS__;scan(__VA_ARGS__)
#define LD(...) ld __VA_ARGS__;scan(__VA_ARGS__)
using namespace std;
using vl=vector<ll>;
using vi=vector<int>;
using vs=vector<string>;
using vc=vector<char>;
using vvl=vector<vl>;
using P=pair<ll,ll>;
using vvc=vector<vc>;
using vd=vector<double>;
using vp=vector<P>;
using vb=vector<bool>;
const int dx[8]={1,0,-1,0,1,-1,-1,1};
const int dy[8]={0,1,0,-1,1,1,-1,-1};
const ll inf=2e18;
const ll MOD=1000000007;
const ll mod=998244353;
const double pi=acos(-1);
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>
ostream &operator<<(ostream&os,const vector<T>&v) {
  for(int i=0;i<(int)v.size();i++) {
    os<<v[i]<<(i+1!=v.size()?" ":"");
  }
  return os;
}
template<typename T>
istream &operator>>(istream&is,vector<T>&v) {
  for(T &in:v)is>>in;
  return is;
}
void scan(){}
template<class Head,class... Tail>
void scan(Head&head,Tail&... tail) {
  cin>>head;
  scan(tail...);
}
template<class T>
void print(const T &t) { cout << t << '\n'; }
template<class Head, class... Tail>
void print(const Head &head, const Tail &... tail) {
  cout << head << ' ';
  print(tail...);
}
template<class... T>
void fin(const T &... a) {
  print(a...);
  exit(0);
}
template<typename T>
ll sum_(vector<T>&v){
  ll res=0;
  for(auto &e:v)res+=e;
  return res;
}
template<typename T1,typename T2>
inline bool chmax(T1&a,T2 b){return a<b&&(a=b,true);}
template<typename T1,typename T2>
inline bool chmin(T1&a,T2 b){return a>b&&(a=b,true);}
#if __has_include(<atcoder/all>)
#include <atcoder/all>
using mint = atcoder::modint1000000007;
#endif
template< typename T >
struct Combination {
  vector< T > _fact, _rfact;

  Combination(int sz) : _fact(sz + 1), _rfact(sz + 1) {
    _fact[0] = _rfact[sz] = 1;
    for(int i = 1; i <= sz; i++) _fact[i] = _fact[i - 1] * i;
    _rfact[sz] /= _fact[sz];
    for(int i = sz - 1; i >= 0; i--) _rfact[i] = _rfact[i + 1] * (i + 1);
  }

  inline T fact(int k) const { return _fact[k]; }

  inline T rfact(int k) const { return _rfact[k]; }

  T P(int n, int r) const {
    if(r < 0 || n < r) return 0;
    return fact(n) * rfact(n - r);
  }

  T C(int p, int q) const {
    if(q < 0 || p < q) return 0;
    return fact(p) * rfact(q) * rfact(p - q);
  }

  T H(int n, int r) const {
    if(n < 0 || r < 0) return (0);
    return r == 0 ? 1 : C(n + r - 1, r);
  }
};
template< typename T >
T lagrange_polynomial(const vector< T > &y, int64_t t) {
  int N=y.size()-1;
  Combination<T>comb(N);
  if(t<=N) return y[t];
  T ret(0);
  vector< T > dp(N + 1, 1), pd(N + 1, 1);
  for(int i = 0; i < N; i++) dp[i + 1] = dp[i] * (t - i);
  for(int i = N; i > 0; i--) pd[i - 1] = pd[i] * (t - i);
  for(int i = 0; i <= N; i++) {
    T tmp = y[i] * dp[i] * pd[i] * comb.rfact(i) * comb.rfact(N - i);
    if((N - i) & 1) ret -= tmp;
    else ret += tmp;
  }
  return ret;
}
long long modpow(long long a,long long n,long long m){
  long long ans=1;
  while(n>0){
    if(n&1)ans=ans*a%m;
    a=a*a%m;
    n>>=1;
  }
  return ans;
}
int main(){
  LL(n,k);
  vector<mint>y(k+2);
  rep(i,1,k+2){
    y[i]=y[i-1];
    y[i]+=modpow(i,k,MOD);
  }
  cout<<lagrange_polynomial(y,n).val()<<endl;
}