STEM Reasoning on JEE Main 2026
97.26Pass@1Qwen3-30B-A3B (Thinking)
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen3-30B-A3B (Thinking)evaluation_protocol=4-sample mean2026.04 | 97.26 | |
| Gemini 2.5 Flashevaluation_protocol=4-sample mean2026.04 | 96.22 | |
| GPT-5 Minievaluation_protocol=4-sample mean2026.04 | 95.83 | |
| GPT-OSS-120Bevaluation_protocol=4-sample mean2026.04 | 95.42 | |
| Nemotron 3 Nano 30B A3Bevaluation_protocol=4-sample mean2026.04 | 94.84 | |
| Aryabhata 2evaluation_protocol=4-sample mean2026.04 | 92.99 | |
| GPT-OSS-20Bevaluation_protocol=4-sample mean2026.04 | 92.46 | |
| GPT-5 Nanoevaluation_protocol=4-sample mean2026.04 | 81.68 |