🔎 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

model: temp:

def f(text):
    ls = list(text)
    for i in range(0, len(ls)):
        if ls[i]!='+':
            ls.insert(i, '+')
            ls.insert(i, '*')
            break
    return '+'.join(ls)
assert f('nzoh') == '*+++n+z+o+h'

CRUXEval-I assert f(??) == '*+++n+z+o+h'

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

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