Tool Calling on BFCL Multiple
92.5AccuracyFull-FT
Evaluation results
| Method | Links | |
|---|---|---|
| Full-FTBackbone=Qwen3-14B-Base, Decode Execution=Independent2026.03 | 92.5 | |
| SUNBackbone=Qwen3-14B-Base, Decode Execution=Shared2026.03 | 92.5 | |
| Full-FTBackbone=Qwen3-8B-Base, Decode Execution=Independent2026.03 | 92 | |
| SUNBackbone=Qwen3-8B-Base, Decode Execution=Shared2026.03 | 91.5 | |
| SUNBackbone=LLaMA3.1-8B, Decode Execution=Shared2026.03 | 90.5 | |
| Full-FTBackbone=Qwen3-1.7B-Base, Decode Execution=Independent2026.03 | 90 | |
| SUNBackbone=Qwen3-1.7B-Base, Decode Execution=Shared2026.03 | 89 | |
| Full-FTBackbone=LLaMA3.1-8B, Decode Execution=Independent2026.03 | 88 | |
| Qwen3-8B-BaseDecode Execution=N/A2026.03 | 80 | |
| Qwen3-14B-BaseDecode Execution=N/A2026.03 | 77 | |
| Qwen3-1.7B-BaseDecode Execution=N/A2026.03 | 60 | |
| LLaMA3.1-8BDecode Execution=N/A2026.03 | 45.5 |