Tool Use on BFCL (test)
90.2AccuracyClaude 3.5 Sonnet
Evaluation results
| Method | Links | |
|---|---|---|
| Claude 3.5 Sonnetevaluation_mode=zero-shot2024.07 | 90.2 | |
| Llama 3 405Bevaluation_mode=zero-shot2024.07 | 88.5 | |
| GPT-4evaluation_mode=zero-shot2024.07 | 88.3 | |
| Nemotron 4 340Bevaluation_mode=zero-shot2024.07 | 86.5 | |
| GPT-3.5 Turboevaluation_mode=zero-shot2024.07 | 85.9 | |
| Llama 3 70Bevaluation_mode=zero-shot2024.07 | 84.8 | |
| GPT-4oevaluation_mode=zero-shot2024.07 | 80.5 | |
| Llama 3 8Bevaluation_mode=zero-shot2024.07 | 76.1 | |
| Mistral 7Bevaluation_mode=zero-shot2024.07 | 60.4 |