🔎 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):
    text = text.upper()
    count_upper = 0
    for char in text:
        if char.isupper():
            count_upper += 1
        else:
            return 'no'
    return count_upper // 2
assert f('ax') == 1
CRUXEval-I 
assert f(??) == 1
 

            		
          

          

            
              CRUXEval-O
              
assert f('ax') == ??

              

              

            		
          

        

      


      
      

      

        
        
        
        

          58 / 800
        

      


      
      

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