ResearchDatasetsHumanEval, MBPPHumanEval, MBPPFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsCodingHumanEval, MBPP50.2HumanEval Score35Code GenerationHumanEval+, MBPP+ v1.0 (test)88.41HumanEval+ Accuracy20