Tool Response Generation on GTM Single-turn
99.6Format ScoreQwen2.5-7B-Instruct
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Qwen2.5-7B-InstructModel Size=7B2025.12 | 99.6 | 99.6 | 99.3 | 99.9 | 98.8 | |
| Qwen2.5-14B-InstructModel Size=14B2025.12 | 99.3 | 99.6 | 99.9 | 99.9 | 98.8 | |
| GTM-1.5BModel Size=1.5B2025.12 | 99.2 | 97.3 | 97.6 | 98 | 95.5 | |
| Qwen2.5-0.5B-InstructModel Size=0.5B2025.12 | 97.4 | 83.4 | 83.8 | 97 | 78.1 | |
| Qwen2.5-1.5B-InstructModel Size=1.5B2025.12 | 97 | 93.1 | 94.8 | 97 | 90.3 | |
| Llama-3.2-3B-InstructModel Size=3B2025.12 | 96.3 | 94.5 | 93.9 | 97.6 | 89.3 | |
| Qwen2.5-3B-InstructModel Size=3B2025.12 | 95.3 | 94.7 | 94.9 | 98.7 | 89.6 | |
| InternLM2.5-20BModel Size=20B2025.12 | 86.4 | 92.6 | 92.8 | 96.4 | 81.8 | |
| InternLM2.5-7BModel Size=7B2025.12 | 72.2 | 70 | 63 | 70.9 | 53.6 | |
| Llama-3.2-1B-InstructModel Size=1B2025.12 | 69.6 | 58.1 | 51.6 | 80.3 | 45.6 | |
| InternLM2.5-1.8BModel Size=1.8B2025.12 | 56 | 8.7 | 5 | 33 | 3 |