Human Proficiency Exam on GMAT
96GMAT Quantitative ScoreLlama 3 405B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Llama 3 405Bprompting=few-shot2024.07 | 96 | 86.6 | |
| GPT-4oprompting=few-shot2024.07 | 92 | 95.5 | |
| Claude 3.5 Sonnetprompting=few-shot2024.07 | 92 | 92.5 | |
| Llama 3 70Bprompting=few-shot2024.07 | 84 | 85.1 | |
| Nemotron 4 340Bprompting=few-shot2024.07 | 76 | — | |
| Llama 3 8Bprompting=few-shot2024.07 | 56 | 65.7 | |
| GPT-3.5 Turboprompting=few-shot2024.07 | 36 | 65.7 |