ResearchDatasetsMCP-BenchMCP-BenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsTool-augmented agent executionMCP-Bench53.5Task Fulfillment32Agentic Tool UseMCP-Bench 3-server 1.03.95Overall Score20Agentic Tool UseMCP-Bench Single 1.03.54Overall Score20Agent Performance EvaluationMCP-Bench46.8Task Fulfillment7End-to-End Question AnsweringMCP-Bench87.5Accuracy (Human)4