AST evaluation on BFCL live_multiple
94.3Free Name AccuracyQwen2.5-7B
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Qwen2.5-7BModel=Qwen2.5-7B2026.06 | 94.3 | 66.7 | 35.3 | 41 | 44.3 | |
| Llama-3.1-8BModel=Llama-3.1-8B2026.06 | 94 | 77.3 | 44.7 | 52 | 54.7 | |
| Pooled (3 main)Model=Pooled (3 main)2026.06 | 92.6 | 57.2 | 32.9 | 38.1 | 40.9 | |
| Qwen3-8BModel=Qwen3-8B2026.06 | 91.3 | 71.7 | 41.7 | 46 | 50.7 | |
| Qwen2.5-1.5BModel=Qwen2.5-1.5B2026.06 | 89.3 | 27.7 | 18.7 | 21.3 | 23.7 | |
| xLAM-2-8B†Model=xLAM-2-8B†2026.06 | 34.7 | 30 | 42.7 | 50.7 | 57 |