ResearchTasksScientific Code GenerationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedSciCode (test)LCP43.8Subset Accuracy36Mar 4, 2026ScienceAgentBench (test)MAA32.8SR32Jun 19, 2026ScienceAgentBenchSciNav25.5SR10Mar 24, 2026Zero-D Reactor combustion code generation——Primary metric0Feb 26, 2026