ResearchTasksTool-use Agent TasksFor AI assistantsMarkdownContent indexTool-use Agent TasksFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedTinyAgent 500 samples (evaluation)Baseline (normal SFT)66.8Accuracy12May 14, 2026