🔎 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(text):
    count = len(text)
    for i in range(-count+1, 0):
        text = text + text[i]
    return text
assert f('wlace A') == 'wlace Alc l  '
CRUXEval-I 
assert f(??) == 'wlace Alc l  '
 

            		
          

          

            
              CRUXEval-O
              
assert f('wlace A') == ??

              

              

            		
          

        

      


      
      

      

        
        
        
        

          251 / 800
        

      


      
      

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