ResearchDatasetsComplexFuncBenchComplexFuncBenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsMulti-Step Tool OrchestrationComplexFuncBench 1.0 (test)36.4Turn Accuracy22Cross-domain ReasoningComplexFuncBench Out-of-Distribution49.7Initial F15Cross-domain ReasoningComplexFuncBench In-Distribution55.6Initial F15