#include<iostream>
#include<cmath>

using namespace std;

#define MAX ((2<<9)+1)
#define INF -1000000000

int bitdp[MAX];
int data[9][9];

void dp(int bit,int n){
	bool f[9];
	for(int i=0;i<n;i++){
		if((bit>>i)%2==0) f[i]=false;
		else f[i]=true;
	}
	for(int i=0;i<n;i++){
		if((bit>>i)%2==1){
			int tmpbit=bit;
			tmpbit-=pow(2.0,i);
			int c=bitdp[tmpbit];
			int pc=0;
			for(int j=0;j<n;j++){
				if(!f[j]) pc+=data[i][j];
			}
			//cout<<bit<<","<<i<<","<<bitdp[bit]<<","<<c<<","<<pc<<endl;
			bitdp[bit]=max(bitdp[bit],c+pc);
		}
	}
}

void make(int n,int c,int bit,int deep,int to,int N){
	if(n==0){
		if(to==c){
			dp(bit,N);
		}
	}
	else{
		if(to!=c) make(n-1,c+1,bit+pow(2.0,deep),deep+1,to,N);
		make(n-1,c,bit,deep+1,to,N);
	}
}

int main(){

	int N,M;
	int item1,item2,score;

	cin>>N>>M;

	for(int i=0;i<M;i++){
		cin>>item1>>item2>>score;
		data[item1][item2]=score;
	}

	for(int i=0;i<=(2<<9);i++) bitdp[i]=INF;
	bitdp[0]=0;

	for(int i=1;i<=N;i++){
		make(N,0,0,0,i,N);
	}

	cout<<bitdp[(int)pow(2.0,N)-1]<<endl;
}