MATH-500: leaderboard, scores and methodology
A 500-problem subset of competition mathematics problems.
Top 10
Bars are scaled to the highest observed score. The leader sets the pace; the spread below it shows how much separation exists in this benchmark.
Leaderboard
| # | Model | Provider | Context | Input / 1M | Avg score |
|---|---|---|---|---|---|
| 1 | GPT-5.2 Max | OpenAI | 1000000 | $21.39 | 89.2 |
| 2 | GPT-5.1 Reasoning | OpenAI | 400000 | $20.27 | 88.3 |
| 3 | GPT-5.1 Pro | OpenAI | 200000 | $19.77 | 87.8 |
| 4 | o4 Reasoning | OpenAI | 1000000 | $19.69 | 87.5 |
| 5 | Grok 4 Max | xAI | 2000000 | $19.70 | 86.6 |
| 6 | Grok 4 Reasoning | xAI | 2000000 | $19.67 | 87.8 |
| 7 | Claude Opus 4.5 Max | Anthropic | 1000000 | $20.90 | 87.8 |
| 8 | Claude Opus 4.5 | Anthropic | 400000 | $19.42 | 87.7 |
| 9 | Gemini 3 Pro Reasoning | Google DeepMind | 200000 | $19.69 | 87.5 |
| 10 | Claude Opus 4.5 Turbo | Anthropic | 2000000 | $19.57 | 85.7 |
| 11 | Qwen3-Max Reasoning | Alibaba Cloud | 2000000 | $18.12 | 85.9 |
| 12 | GPT-5.2 Vision | OpenAI | 1000000 | $19.03 | 87.4 |
| 13 | Claude Sonnet 4.5 Max | Anthropic | 1000000 | $19.90 | 87.2 |
| 14 | Qwen3-Max Max | Alibaba Cloud | 400000 | $18.53 | 85.9 |
| 15 | GPT-5.2 Turbo | OpenAI | 1000000 | $20.73 | 87.4 |
| 16 | Gemini 3 Pro Turbo | Google DeepMind | 400000 | $19.62 | 86.4 |
| 17 | o3 Pro | OpenAI | 400000 | $17.89 | 85.8 |
| 18 | Grok 4 Pro | xAI | 1000000 | $19.00 | 85.4 |
| 19 | Gemini 3 Pro Max | Google DeepMind | 2000000 | $20.57 | 87.9 |
| 20 | Grok 4 Turbo | xAI | 1000000 | $19.21 | 86.7 |
| 21 | o3 Reasoning | OpenAI | 200000 | $18.07 | 85.3 |
| 22 | Claude Opus 4.5 Vision | Anthropic | 200000 | $17.70 | 85.9 |
| 23 | Grok 3 Max | xAI | 2000000 | $17.66 | 85.9 |
| 24 | Qwen3-Max Turbo | Alibaba Cloud | 1000000 | $17.70 | 84.7 |
| 25 | GPT-5.2 Pro | OpenAI | 2000000 | $21.32 | 89.6 |
| 26 | Gemini 3 Flash Pro | Google DeepMind | 400000 | $17.71 | 84.8 |
| 27 | Qwen3-Max Pro | Alibaba Cloud | 400000 | $18.87 | 86.2 |
| 28 | o4 Turbo | OpenAI | 2000000 | $19.34 | 86.6 |
| 29 | o4 Vision | OpenAI | 200000 | $17.01 | 84.5 |
| 30 | Claude Opus 4.5 Reasoning | Anthropic | 2000000 | $20.91 | 88.2 |
| 31 | Gemini 2.5 Pro Max | Google DeepMind | 200000 | $16.94 | 84.8 |
| 32 | Gemini 3 Pro Pro | Google DeepMind | 200000 | $19.80 | 87.3 |
| 33 | GPT-5.1 Turbo | OpenAI | 2000000 | $19.19 | 86.1 |
| 34 | GPT-5.2 Reasoning | OpenAI | 1000000 | $20.98 | 87.5 |
| 35 | GPT-5.1 Max | OpenAI | 2000000 | $20.50 | 87.9 |
| 36 | Claude Opus 4.5 Pro | Anthropic | 200000 | $20.98 | 89.1 |
| 37 | Grok 4 | xAI | 200000 | $17.90 | 86.5 |
| 38 | Claude Sonnet 4.5 Vision | Anthropic | 128000 | $16.99 | 83.8 |
| 39 | GPT-5.2 | OpenAI | 2000000 | $19.71 | 88.0 |
| 40 | Gemini 2.5 Pro Turbo | Google DeepMind | 1000000 | $16.32 | 83.9 |
| 41 | o3 Turbo | OpenAI | 200000 | $16.63 | 83.4 |
| 42 | o4 Max | OpenAI | 2000000 | $19.51 | 86.2 |
| 43 | GPT-5.1 | OpenAI | 2000000 | $19.01 | 86.0 |
| 44 | o4 | OpenAI | 2000000 | $17.74 | 85.4 |
| 45 | Claude 3.7 Sonnet Reasoning | Anthropic | 128000 | $16.30 | 83.7 |
| 46 | DeepSeek-V3.2 Chat | DeepSeek | 1000000 | $5.80 | 82.5 |
| 47 | Grok 3 Reasoning | xAI | 200000 | $16.07 | 82.9 |
| 48 | Claude Sonnet 4.5 Pro | Anthropic | 1000000 | $18.73 | 86.7 |
| 49 | Gemini 3 Pro | Google DeepMind | 2000000 | $18.24 | 86.1 |
| 50 | o4 Pro | OpenAI | 1000000 | $19.37 | 87.3 |
Methodology
MATH-500 is included as a mathematics benchmark with a typical score range of 0 to 100. The values in this index are deterministic modelled estimates generated for comparison at scale. They are not official leaderboard results and should not be quoted as vendor benchmarks.
Each model receives a score derived from its capability tier, release generation and benchmark difficulty. Scores are normalised to a 0–100 scale so that models from different providers can be placed side by side. The average shown is the arithmetic mean across all 600 scored models.
Read the leaderboard as a relative map. A five-point gap is meaningful at the top and less meaningful in the middle of the distribution. Pair the score with price and context data before choosing a model.
How to use this leaderboard
- Use the top 10 to shortlist candidates, not to declare a winner.
- Check whether the leader is open weights or proprietary before assuming you can deploy it.
- Compare the score against input and output price on each model page.
- Look at the model's scores on related benchmarks before trusting a single number.
FAQ
What is MATH-500?
A 500-problem subset of competition mathematics problems.
Who leads MATH-500?
GPT-5.2 Max from OpenAI leads this index with a score of 96.7, followed by GPT-5.1 Reasoning at 95.9.
What is a good MATH-500 score?
The top quartile starts around 82.8. The index average is 72.5, so a score above 82.8 is a strong result relative to this dataset.
How are MATH-500 scores calculated?
The values are modelled estimates on a 0–100 scale, generated from capability tier, release generation and benchmark difficulty. They are not official leaderboard results.
Which models are included?
All 600 models in the index are scored on MATH-500, across 40 providers and every model category.