ResearchTasksCode-generation monitoringCode-generation monitoringFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedEvaluation objectives Standard prompt setting (out-of-domain)Gemini-3-Flash100Exit Rate22May 14, 2026
Evaluation objectives Standard prompt setting (out-of-domain)Gemini-3-Flash100Exit Rate22May 14, 2026