Tool Usage on BFCL Multi-Parallel v1
90.5AccuracyQwen3-4B-Thinking
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Qwen3-4B-ThinkingBase Model=Qwen3-4B-Thinking, Stopping Strategy=Base2025.10 | 90.5 | — | |
| THINKBRAKEBase Model=Qwen3-4B-Thinking, Stopping Strategy=THINKBRAKE2025.10 | 90.5 | -28.2 | |
| THINKBRAKEBase Model=Qwen3-32B, Stopping Strategy=THINKBRAKE2025.10 | 90 | -7.3 | |
| NoWaitBase Model=Qwen3-4B-Thinking, Stopping Strategy=NoWait2025.10 | 88.5 | -53.5 | |
| THINKBRAKEBase Model=Qwen3-14B, Stopping Strategy=THINKBRAKE2025.10 | 85 | -14.2 | |
| Qwen3-32BBase Model=Qwen3-32B, Stopping Strategy=Base2025.10 | 85 | — | |
| NoWaitBase Model=Qwen3-4B, Stopping Strategy=NoWait2025.10 | 84 | 1.2 | |
| NoWaitBase Model=Qwen3-14B, Stopping Strategy=NoWait2025.10 | 84 | 1.8 | |
| THINKBRAKEBase Model=Qwen3-4B, Stopping Strategy=THINKBRAKE2025.10 | 83.5 | -12 | |
| Qwen3-14BBase Model=Qwen3-14B, Stopping Strategy=Base2025.10 | 83.5 | — | |
| ThinkLessBase Model=Qwen3-4B-Thinking, Stopping Strategy=ThinkLess2025.10 | 83 | -100 | |
| Qwen3-4BBase Model=Qwen3-4B, Stopping Strategy=Base2025.10 | 83 | — | |
| NoWaitBase Model=Qwen3-32B, Stopping Strategy=NoWait2025.10 | 82.5 | -5 | |
| Phi-4-ReasoningBase Model=Phi-4-Reasoning, Stopping Strategy=Base2025.10 | 77.5 | — | |
| THINKBRAKEBase Model=Phi-4-Reasoning, Stopping Strategy=THINKBRAKE2025.10 | 77.5 | -6.4 | |
| NoWaitBase Model=Phi-4-Reasoning, Stopping Strategy=NoWait2025.10 | 75.5 | -13.2 | |
| THINKBRAKE-pBase Model=Qwen3-4B-Thinking, Stopping Strategy=THINKBRAKE-p2025.10 | 42.5 | -33.4 | |
| NoWaitBase Model=DeepSeek-R1-7B, Stopping Strategy=NoWait2025.10 | 42.5 | -6.9 | |
| DeepSeek-R1-7BBase Model=DeepSeek-R1-7B, Stopping Strategy=Base2025.10 | 38 | — | |
| THINKBRAKEBase Model=DeepSeek-R1-7B, Stopping Strategy=THINKBRAKE2025.10 | 37 | -11.6 |