Tool selection on ToolAlpaca
97.42AccuracyMTA
Evaluation Results
| Method | Links | |
|---|---|---|
| MTABackbone=Qwen1.5-Chat (7B)2026.01 | 97.42 | |
| FTBackbone=Qwen1.5-Chat (7B)2026.01 | 96.88 | |
| MTABackbone=Baichuan2-Chat (7B)2026.01 | 95.7 | |
| FTBackbone=Baichuan2-Chat (7B)2026.01 | 95.12 | |
| MTABackbone=Chinese Alpaca-2 (7B)2026.01 | 91.03 | |
| FTBackbone=Chinese Alpaca-2 (7B)2026.01 | 90.12 | |
| MTABackbone=ChatGLM3 (6B)2026.01 | 84.02 | |
| FTBackbone=ChatGLM3 (6B)2026.01 | 83.56 | |
| ReActBackbone=ChatGLM3 (6B)2026.01 | 68.21 | |
| ReActBackbone=Qwen1.5-Chat (7B)2026.01 | 62.11 | |
| CoTBackbone=ChatGLM3 (6B)2026.01 | 58.45 | |
| BaseBackbone=ChatGLM3 (6B)2026.01 | 53.89 | |
| ReActBackbone=Baichuan2-Chat (7B)2026.01 | 53.45 | |
| CoTBackbone=Qwen1.5-Chat (7B)2026.01 | 51.89 | |
| CoTBackbone=Baichuan2-Chat (7B)2026.01 | 48.45 | |
| BaseBackbone=Qwen1.5-Chat (7B)2026.01 | 47.66 | |
| ReActBackbone=Chinese Alpaca-2 (7B)2026.01 | 44.39 | |
| BaseBackbone=Baichuan2-Chat (7B)2026.01 | 44.23 | |
| CoTBackbone=Chinese Alpaca-2 (7B)2026.01 | 40.88 | |
| BaseBackbone=Chinese Alpaca-2 (7B)2026.01 | 36.23 |