Mathematics Problem Solving
Competition-level mathematics problems spanning algebra, geometry, number theory, and calculus. Tests multi-step mathematical reasoning.
| # | Model | Score | |
|---|---|---|---|
| 1 | GPT-5.2OpenAI | 98% | Try |
| 2 | Claude Sonnet 4.6Anthropic | 97.8% | Try |
| 3 | Claude Opus 4.6Anthropic | 97.6% | Try |
| 4 | DeepSeek R1DeepSeek | 97.3% | Try |
| 5 | Gemini 3.1 ProGoogle | 96.8% | Try |
| 6 | GPT-5.3 CodexOpenAI | 96% | Try |
| 7 | Gemini 3 ProGoogle | 95% | Try |
| 8 | o4-miniOpenAI | 93.4% | Try |
| 9 | Claude Opus 4.5Anthropic | 92% | Try |
| 10 | Grok 4xAI | 91.7% | Try |
| 11 | o3OpenAI | 91.6% | Try |
| 12 | Gemini 2.5 ProGoogle | 90.2% | Try |
| 13 | Gemini 3 FlashGoogle | 90% | Try |
| 14 | Claude Sonnet 4.5Anthropic | 87% | Try |
| 15 | Gemini 2.5 FlashGoogle | 82.1% | Try |
| 16 | Llama 4 MaverickMeta | 75.8% | Try |
| 17 | GPT-4.1OpenAI | 73.8% | Try |
| 18 | Llama 4 ScoutMeta | 70.1% | Try |