ResearchTasksAgent & OpenClawFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedPinchBenchQwen3-Next-80B-A3B83.7Accuracy5Apr 6, 2026τ²-BenchQwen3.5-35B-A3B76.6Accuracy5Apr 6, 2026