# AA MATH-500 leaderboard

Mathematics benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/math-aa-math500

Artificial Analysis MATH-500. Every model the catalog carries a published AA MATH-500 value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | Mathematics |
| Measure | Accuracy |
| Tasks | 500 competition mathematics problems |
| Difficulty | High school to undergraduate mathematics |
| Published by | [Artificial Analysis MATH-500 Benchmark Leaderboard](https://artificialanalysis.ai/evaluations/math-500) |

An independently evaluated MATH-500 result from Artificial Analysis.

## AA MATH-500 ranking

11 models with a published AA MATH-500 value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Accuracy | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [GPT-5](https://modelchoice.dev/models/gpt-5) | OpenAI | 99.4 | Observed 2026-11-21 · source benchlm:models |
| 1 | [GPT-5 (high)](https://modelchoice.dev/models/gpt-5-high) | OpenAI | 99.4 | Last good 2026-11-18 · source benchlm:models |
| 3 | [o3](https://modelchoice.dev/models/o3) | OpenAI | 99.2 | Observed 2026-11-21 · source benchlm:models |
| 4 | [GPT-5 (medium)](https://modelchoice.dev/models/gpt-5-medium) | OpenAI | 99.1 | Last good 2026-10-23 · source benchlm:models |
| 5 | [Grok 4](https://modelchoice.dev/models/grok-4) | xAI | 99.0 | Observed 2026-11-21 · source benchlm:models |
| 6 | [o4-mini](https://modelchoice.dev/models/o4-mini) | OpenAI | 98.9 | Observed 2026-11-21 · source benchlm:models |
| 7 | [DeepSeek-R1](https://modelchoice.dev/models/deepseek-r1) | DeepSeek | 98.3 | Observed 2026-11-21 · source benchlm:models |
| 8 | [Claude 4 Opus Thinking](https://modelchoice.dev/models/claude-4-opus-thinking) | Anthropic | 98.2 | Observed 2026-11-21 · source benchlm:models |
| 9 | [Kimi K2](https://modelchoice.dev/models/kimi-k2) | Moonshot AI | 97.1 | Observed 2026-11-21 · source benchlm:models |
| 10 | [o1](https://modelchoice.dev/models/o1) | OpenAI | 97.0 | Observed 2026-11-21 · source benchlm:models |
| 11 | [Gemini 2.5 Pro](https://modelchoice.dev/models/gemini-2-5-pro) | Google | 96.7 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
