ResearchTasksAgentic ReasoningFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedTIR-BenchPyVision-Image19.8Accuracy3Feb 26, 2026ZebraArena multi-turn LargeCopT82.24Accuracy2May 20, 2026ZebraArena multi-turn MediumCopT88.14Accuracy2May 20, 2026ZebraArena multi-turn (Small)CopT96.69Accuracy2May 20, 2026TauBench Telecom V2Nemotron 3 Nano Omni42.7Accuracy2Apr 29, 2026τ-Bench (test)——ScoreAgentic Reasoning on τ-Bench (test) performance trend0Feb 26, 2026Page 2 of 2PreviousNext