HumanEval

github.com ↗

- HumanEval is a benchmark for evaluating the functional correctness of code generation models using Python programming problems with unit tests.