ResearchDatasetsHEvalHEvalFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsCode generationHEval68.29Pass@121CodingHEval84.8Accuracy20Instruction-followingHEval46.12PASS@112CodingHEval+75Accuracy12