Mathematical Reasoning on MathCheck (129-group)
95.3SCR+Sonnet
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Sonnetevaluation_mode=canonical + PU2026.05 | 95.3 | 98.4 | 0 | |
| GPT-4oevaluation_mode=canonical + PU2026.05 | 94.6 | 95.2 | 2 | |
| Haikuevaluation_mode=canonical + PU2026.05 | 93.8 | 96.8 | 1 | |
| DeepSeek V3evaluation_mode=canonical + PU2026.05 | 93 | 96 | 1 |