ResearchBenchmarksFunction Calling on APIGen (test)Follow89.6Score (Single)Ours89.0889.21589.3589.485Jan 25, 2026Evaluation ResultsMethodMethodLinksScore (Single)Score (Parallel)Score (None)Score (All Modes)OursFine-tuning dataset=Th...Fine-tuning dataset=This paper's dataset2026.0189.685.831.273.2ToolAceFine-tuning dataset=To...Fine-tuning dataset=ToolAce2026.0189.170.728.569