Mathematical Problem Solving on Olympiad Bench (Pass@1)
70.36Pass@1 AccuracyQUESTA-Nemotron-1.5B
Evaluation Results
| Method | Links | |
|---|---|---|
| QUESTA-Nemotron-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 70.36 | |
| Qwen3-8Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 68.56 | |
| DeepSeek-R1-Distill-32Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 65 | |
| Nemotron-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 64.62 | |
| Qwen3-1.7Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 56.13 | |
| DeepSeek-R1-Distill-1.5Bk (responses per question)=32, sampling temperature=0.7, top-p=0.952025.07 | 52.4 |