🔎 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(mylist):
    revl = mylist[:]
    revl.reverse()
    mylist.sort(reverse=True)
    return mylist == revl
assert f([5, 8]) == True
CRUXEval-I 
assert f(??) == True
 

            		
          

          

            
              CRUXEval-O
              
assert f([5, 8]) == ??

              

              

            		
          

        

      


      
      

      

        
        
        
        

          29 / 800
        

      


      
      

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