#pragma GCC target("avx2")
#pragma GCC optimize("O3")
#pragma GCC optimize("unroll-loops")
#include <bits/stdc++.h>
#define MOD 998244353

int ri() {
	int n;
	scanf("%d", &n);
	return n;
}
int main() {
	int n = ri();
	int k = ri();
	int kk = k * k;
	int kkk = k * k * k;
	int a[kkk][kkk];
	memset(a, 0, sizeof(a));
	for (int i = 0; i < k*k*k; i++) {
		int x = i / kk;
		int y = i / k % k;
		int z = i % k;
		a[i][x * kk + y * k + (z + 1) % k]++;
		a[i][x * kk + (y + z) % k * k + z]++;
		a[i][(x + y) % k * kk + y * k + z]++;
	}
	std::vector<int> res(kkk, 0);
	std::vector<int> tmp(kkk);
	res[0] = 1;
	int kkk_1 = (kkk&0x7FFFFFF0);
	int kkk_f = (kkk&0xF);
	while (n) {
		if (n & 1) {
			tmp.assign(kkk, 0);
			for (int i = 0; i < kkk; i++)
				for (int j = 0; j < kkk; j++) {
					tmp[j] += (int64_t) res[i] * a[i][j] % MOD;
					if (tmp[j] >= MOD) tmp[j] -= MOD;
				}
			std::swap(tmp, res);
		}
		int b[kkk][kkk];
		memset(b, 0, sizeof(b));
		for (int i = 0; i < kkk; i++) {
			for (int j = 0; j < kkk; j++) {
				uint64_t sum = 0;
				for (int l = 0; l < kkk_1; l += 0x10) {
					sum += (uint64_t) a[i][l] * a[l][j];
					sum += (uint64_t) a[i][l + 1] * a[l + 1][j];
					sum += (uint64_t) a[i][l + 2] * a[l + 2][j];
					sum += (uint64_t) a[i][l + 3] * a[l + 3][j];
					sum += (uint64_t) a[i][l + 4] * a[l + 4][j];
					sum += (uint64_t) a[i][l + 5] * a[l + 5][j];
					sum += (uint64_t) a[i][l + 6] * a[l + 6][j];
					sum += (uint64_t) a[i][l + 7] * a[l + 7][j];
					sum += (uint64_t) a[i][l + 8] * a[l + 8][j];
					sum += (uint64_t) a[i][l + 9] * a[l + 9][j];
					sum += (uint64_t) a[i][l + 10] * a[l + 10][j];
					sum += (uint64_t) a[i][l + 11] * a[l + 11][j];
					sum += (uint64_t) a[i][l + 12] * a[l + 12][j];
					sum += (uint64_t) a[i][l + 13] * a[l + 13][j];
					sum += (uint64_t) a[i][l + 14] * a[l + 14][j];
					sum += (uint64_t) a[i][l + 15] * a[l + 15][j];
					sum %= MOD;
				}
				for (int l = 0; l < kkk_f; l++) {
					sum += (uint64_t) a[i][kkk_1 + l] * a[kkk_1 + l][j];
				}
				b[i][j] = sum % MOD;
			}
		}
		memcpy(a, b, sizeof(a));
		n /= 2;
	}
	int ans = 0;
	for (int i = 0; i < kk; i++) {
		ans += res[i];
		if (ans >= MOD) ans -= MOD;
	}
	std::cout << ans << std::endl;
	return 0;
}