Math Benchmarks
Mathematical reasoning and problem solving - Compare AI models across 7 mathematical benchmarks including AIME, HMMT, BRUMO, and more.
Filters & Search
Filter models by creator or search by name to find the perfect AI model for your needs
Math Benchmark Results
Showing 15 of 20 models • Click column headers to sort
AIME 2023High school mathematics competition | AIME 2024High school mathematics competition | AIME 2025High school mathematics competition | HMMT Feb 2023Collegiate mathematics competition | HMMT Feb 2024Collegiate mathematics competition | HMMT Feb 2025Collegiate mathematics competition | BRUMO 2025University-level mathematics olympiad | ||||
|---|---|---|---|---|---|---|---|---|---|---|
1 GPT-5 (high) OpenAI | OpenAI | Closed-source | 69 | 93% | 95% | 94% | 88% | 90% | 89% | 91% |
2 GPT-5 (medium) OpenAI | OpenAI | Closed-source | 68 | 91% | 93% | 92% | 86% | 88% | 87% | 89% |
3 Grok 4 xAI | xAI | Closed-source | 68 | 86% | 88% | 87% | 83% | 85% | 84% | 86% |
4 o3-pro OpenAI | OpenAI | Closed-source | 68 | 89% | 91% | 90% | 85% | 87% | 86% | 88% |
5 o3 OpenAI | OpenAI | Closed-source | 67 | 87% | 89% | 88% | 83% | 85% | 84% | 86% |
6 o4-mini (high) OpenAI | OpenAI | Closed-source | 65 | 83% | 85% | 84% | 79% | 81% | 80% | 82% |
7 Gemini 2.5 Pro Google | Closed-source | 65 | 84% | 86% | 85% | 80% | 82% | 81% | 83% | |
8 GPT-5 mini OpenAI | OpenAI | Closed-source | 64 | 80% | 82% | 81% | 76% | 78% | 77% | 79% |
9 Claude 4.1 Opus Anthropic | Anthropic | Closed-source | 61 | 76% | 78% | 77% | 72% | 74% | 73% | 75% |
10 Claude 4 Sonnet Anthropic | Anthropic | Closed-source | 59 | 73% | 75% | 74% | 69% | 71% | 70% | 72% |
11 Llama 3.1 405B Meta | Meta | Open-source | 58 | 70% | 72% | 71% | 66% | 68% | 67% | 69% |
12 Mistral Large 2 Mistral | Mistral | Open-source | 57 | 68% | 70% | 69% | 64% | 66% | 65% | 67% |
13 GPT-4o OpenAI | OpenAI | Closed-source | 56 | 66% | 68% | 67% | 62% | 64% | 63% | 65% |
14 Claude 3.5 Sonnet Anthropic | Anthropic | Closed-source | 55 | 65% | 67% | 66% | 61% | 63% | 62% | 64% |
15 Gemini 1.5 Pro Google | Closed-source | 54 | 64% | 66% | 65% | 60% | 62% | 61% | 63% |
Showing 15 of 20 models
About Math Benchmarks
AIME 2023
High school mathematics competition
AIME 2024
High school mathematics competition
AIME 2025
High school mathematics competition
HMMT Feb 2023
Collegiate mathematics competition
HMMT Feb 2024
Collegiate mathematics competition
HMMT Feb 2025
Collegiate mathematics competition
BRUMO 2025
University-level mathematics olympiad