🔎 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(nums):
    for i in range(len(nums)):
        nums.insert(i, nums[i]**2)
    return nums
assert f([1, 2, 4]) == [1, 1, 1, 1, 2, 4]
CRUXEval-I 
assert f(??) == [1, 1, 1, 1, 2, 4]
 

            		
          

          

            
              CRUXEval-O
              
assert f([1, 2, 4]) == ??

              

              

            		
          

        

      


      
      

      

        
        
        
        

          551 / 800
        

      


      
      

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