Reasoning on MATH-500, AIME 24, AIME 25, GPQA Diamond, CommonsenseQA, LiveCodeBench, and LongBenchv2 Qwen3
74.8Accuracy
Base Model
Updated 4d ago
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
2026.01 | 74.8 | - | 69.2 | |
2026.01 | 72.2 | - | 72.3 | |
2026.01 | 72.1 | - | 72.5 | |
2026.01 | 69.9 | - | 71.6 | |
2026.01 | 69 | - | 75.5 | |
2026.01 | 68.7 | - | 76.6 | |
2026.01 | 50.9 | - | 65.4 | |
2026.01 | 50.7 | - | 59.7 | |
2026.01 | 50.4 | 513,318 | 59.6 | |
2026.01 | 50.3 | - | 62.1 | |
2026.01 | 49.1 | - | 62.1 | |
2026.01 | 48.3 | - | 62.1 |