#include <bits/stdc++.h>
#define int int64_t
const int inf = 1e18;
using namespace std;
#define ft front()
#define bk back()
#define pb push_back
#define ins insert
#define lb lower_bound
#define ub upper_bound
#define all(x) begin(x), end(x)
#define len(x) int(size(x))
#define pct __builtin_popcount
#define clz __builtin_clz
int bits(int x) {
return 63 - clz(x);
}
int p2(int x) {
return int(1) << x;
}
template<class T>
bool ckmax(T& a, const T& b) {
return a < b ? a = b, 1 : 0;
}
template<class T>
bool ckmin(T& a, const T& b) {
return a > b ? a = b, 1 : 0;
}
template<class T> using V = vector<T>;
using vi = V<int>;
using vb = V<bool>;
using ai2 = array<int, 2>;
using ai3 = array<int, 3>;
using ai4 = array<int, 4>;
template<class T, class U>
void erase(T& aa, const U& x) {
auto it = aa.find(x);
assert(it != end(aa));
aa.erase(it);
}
void ps() { cerr << '\n'; }
template<class T, class... Ts> void ps(const T& x, const Ts&... xs) {
cerr << x;
if (sizeof...(xs))
cerr << ' ';
ps(xs...);
}
const int P = 1e9 + 7;
void add(int& a, int b) {
a += b;
if (a < 0)
a += P;
if (a >= P)
a -= P;
}
int fadd(int a, int b) {
add(a, b);
return a;
}
void mul(int& a, int b) {
a = int64_t(a) * b % P;
}
int fmul(int a, int b) {
mul(a, b);
return a;
}
const int N = 100, L = 1000;
int dp[N + 1][N + 1][L + 1][3];
void solve() {
int n, max_cost;
cin >> n >> max_cost;
vi aa(n);
for (auto& x : aa)
cin >> x;
sort(all(aa));
if (n == 1) {
cout << 1 << '\n';
return;
}
// dp[i][j][k][l] = # permutations which
// - processed the first i values
// - have j components
// - have cost k if ? were replaced with aa[i]
// - filled in l <= 2 of the edge values
dp[0][0][0][0] = 1;
for (int i = 0; i < n; i++) {
for (int j = 0; j <= i; j++) {
for (int k = 0; k <= max_cost; k++) {
for (int l = 0; l <= 2; l++) if (dp[i][j][k][l]) {
if (l == 2)
assert(j > 1);
auto cnt_qm_blocks = j + 1 - l;
auto cnt_edges = 2 * j - l;
// create new cc
{
int dk = 0;
if (i + 1 < n)
dk += (aa[i + 1] - aa[i]) * (cnt_edges + 2);
if (k + dk <= max_cost)
add(dp[i + 1][j + 1][k + dk][l], fmul(j + 1 - l, dp[i][j][k][l]));
}
// create new cc (side)
if (l < 2) {
int dk = 0;
if (i + 1 < n)
dk += (aa[i + 1] - aa[i]) * (cnt_edges + 1);
if (k + dk <= max_cost)
add(dp[i + 1][j + 1][k + dk][l + 1], fmul(2 - l, dp[i][j][k][l]));
}
// merge two ccs
if (j > 1 && !(l == 2 && j == 2 && i < n - 1)) {
int dk = 0;
if (i + 1 < n)
dk += (aa[i + 1] - aa[i]) * (cnt_edges - 2);
add(dp[i + 1][j - 1][k + dk][l], fmul(j - 1, dp[i][j][k][l]));
}
// glue to one side of a cc
if (j) {
int dk = 0;
if (i + 1 < n)
dk += (aa[i + 1] - aa[i]) * cnt_edges;
add(dp[i + 1][j][k + dk][l], fmul(2 * j - l, dp[i][j][k][l]));
}
// glue to one side of a cc and become side
if (j && l < 2 && !(l == 1 && j == 1 && i < n - 1)) {
int dk = 0;
assert(cnt_edges);
if (i + 1 < n)
dk += (aa[i + 1] - aa[i]) * (cnt_edges - 1);
add(dp[i + 1][j][k + dk][l + 1], fmul(2 - l, dp[i][j][k][l]));
}
}
}
}
}
int ans = 0;
for (int k = 0; k <= max_cost; k++) {
// ps(k, dp[n][1][k][2]);
add(ans, dp[n][1][k][2]);
}
cout << ans << '\n';
}
signed main() {
cin.tie(0);
cout.tie(0);
ios::sync_with_stdio(0);
solve();
}
