List your company
ResearchFormerly WizWand

Tool Usage on Aggregate BFCL and Meta-Tool

86.6AccuracyTHINKBRAKE
44.89655.72366.5577.377Oct 1, 2025
Updated 6mo ago

Evaluation results

MethodLinks
THINKBRAKEBase Model=Qwen3-4B-Thinking, Stopping Strategy=THINKBRAKE2025.1086.6-22.6
Qwen3-4B-ThinkingBase Model=Qwen3-4B-Thinking, Stopping Strategy=Base2025.1084.3—
Qwen3-4BBase Model=Qwen3-4B, Stopping Strategy=Base2025.1081.6—
NoWaitBase Model=Qwen3-4B-Thinking, Stopping Strategy=NoWait2025.1081.2-37.7
NoWaitBase Model=Qwen3-4B, Stopping Strategy=NoWait2025.1080.8-6.7
THINKBRAKEBase Model=Phi-4-Reasoning, Stopping Strategy=THINKBRAKE2025.1079.2-2
THINKBRAKEBase Model=Qwen3-4B, Stopping Strategy=THINKBRAKE2025.1079-10.7
Phi-4-ReasoningBase Model=Phi-4-Reasoning, Stopping Strategy=Base2025.1078.8—
Qwen3-32BBase Model=Qwen3-32B, Stopping Strategy=Base2025.1077.8—
NoWaitBase Model=Phi-4-Reasoning, Stopping Strategy=NoWait2025.1077.2-9.5
THINKBRAKEBase Model=Qwen3-32B, Stopping Strategy=THINKBRAKE2025.1076.3-7
Qwen3-14BBase Model=Qwen3-14B, Stopping Strategy=Base2025.1076.2—
THINKBRAKEBase Model=Qwen3-14B, Stopping Strategy=THINKBRAKE2025.1075.8-12.4
NoWaitBase Model=Qwen3-32B, Stopping Strategy=NoWait2025.1075.4-5.2
NoWaitBase Model=Qwen3-14B, Stopping Strategy=NoWait2025.1074.1-10.6
ThinkLessBase Model=Qwen3-4B-Thinking, Stopping Strategy=ThinkLess2025.1066.6-100
THINKBRAKE-pBase Model=Qwen3-4B-Thinking, Stopping Strategy=THINKBRAKE-p2025.1061.6-28.5
NoWaitBase Model=DeepSeek-R1-7B, Stopping Strategy=NoWait2025.1052.2-4.6
DeepSeek-R1-7BBase Model=DeepSeek-R1-7B, Stopping Strategy=Base2025.1051.2—
THINKBRAKEBase Model=DeepSeek-R1-7B, Stopping Strategy=THINKBRAKE2025.1046.5-14.3