Tool Use on API Bank
90AccuracyLlama 3.1 Instruct
Evaluation Results
| Method | Links | |
|---|---|---|
| Llama 3.1 InstructModel Scale=70B2025.04 | 90 | |
| Llama 3 InstructModel Scale=70B2025.04 | 85.2 | |
| ParamΔModel Scale=70B2025.04 | 82.9 | |
| Llama 3.1 InstructModel Scale=8B2025.04 | 82.1 | |
| ParamΔModel Scale=8B2025.04 | 51.9 | |
| Llama 3 InstructModel Scale=8B2025.04 | 48.9 | |
| Llama 3 BaseModel Scale=70B2025.04 | 37.9 | |
| Llama 3 BaseModel Scale=8B2025.04 | 25.3 | |
| Llama 3.1 BaseModel Scale=8B2025.04 | 24.8 | |
| Llama 3.1 BaseModel Scale=70B2025.04 | 13.3 |