Models Tested
8
Best Score
93.3%
Median Score
86.15%
Scoring: accuracy
Introduced: 2025-02
Maintainer: Mathematical Association of America
Leaderboard (8 models)
| # | Model | Developer | Score |
|---|---|---|---|
| 🥇 | Grok-3 | xAI | 93.3% |
| 🥈 | o4-mini | OpenAI | 92.6% |
| 🥉 | Claude Sonnet 4.5 | Anthropic | 87% |
| 4 | o3-mini | OpenAI | 86.5% |
| 5 | Gemini 2.5 Pro | Google DeepMind | 85.8% |
| 6 | o3 | OpenAI | 85.3% |
| 7 | DeepSeek R1 | DeepSeek | 74% |
| 8 | Gemini 2.5 Flash | Google DeepMind | 72% |