ResearchDatasetsTIR-BenchFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsMultimodal Agent TaskTIR-Bench47.75Average@424Multimodal Tool-UseTIR-Bench50Avg@416Visual ReasoningTIR-Bench51.8Average Score15Visual NavigationTIR-Bench Maze65Accuracy9Tool-Integrated ReasoningTIR-Bench20.8Score4Agentic ReasoningTIR-Bench19.8Accuracy3