🔎 CRUXEval Sample Explorer 🔎

CRUXEval is a benchmark complementary to HumanEval and MBPP measuring code reasoning, understanding, and execution capabilities!

Homepage Paper Code HF Dataset Leaderboard

    
def f(num):
    if 0 < num < 1000 and num != 6174:
        return 'Half Life'
    return 'Not found'
assert f(6173) == 'Not found'
CRUXEval-I 
assert f(??) == 'Not found'
 

            		
          

          

            
              CRUXEval-O
              
assert f(6173) == ??

              

              

            		
          

        

      


      
      

      

        
        
        
        

          298 / 800
        

      


      
      

        The site is inspired by and based on
        Minerva sample explorer and
        Llemma. Thanks!