> ## Documentation Index
> Fetch the complete documentation index at: https://openbench.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Math

> Evaluate AI models on mathematical reasoning and problem solving

openbench includes a comprehensive collection of math benchmarks ranging from multilingual grade school problems to competition-level mathematics from prestigious contests like AIME and HMMT.

## Available Benchmarks

<CardGroup cols={2}>
  <Card title="AIME 2023" icon="calculator">
    American Invitational Mathematics Examination 2023 - challenging high school competition problems.

    ```bash theme={null}
    bench eval aime_2023_I
    bench eval aime_2023_II
    ```
  </Card>

  <Card title="AIME 2024" icon="calculator">
    American Invitational Mathematics Examination 2024.

    ```bash theme={null}
    bench eval aime_2024
    bench eval aime_2024_I
    bench eval aime_2024_II
    ```
  </Card>

  <Card title="AIME 2025" icon="calculator">
    American Invitational Mathematics Examination 2025.

    ```bash theme={null}
    bench eval aime_2025
    bench eval aime_2025_II
    ```
  </Card>

  <Card title="HMMT Feb 2023" icon="trophy">
    Harvard-MIT Mathematics Tournament February 2023.

    ```bash theme={null}
    bench eval hmmt_feb_2023
    ```
  </Card>

  <Card title="HMMT Feb 2024" icon="trophy">
    Harvard-MIT Mathematics Tournament February 2024.

    ```bash theme={null}
    bench eval hmmt_feb_2024
    ```
  </Card>

  <Card title="HMMT Feb 2025" icon="trophy">
    Harvard-MIT Mathematics Tournament February 2025.

    ```bash theme={null}
    bench eval hmmt_feb_2025
    ```
  </Card>

  <Card title="BRUMO 2025" icon="award">
    BRUMO mathematics competition 2025.

    ```bash theme={null}
    bench eval brumo_2025
    ```
  </Card>

  <Card title="MATH" icon="function">
    Competition-level math problems covering algebra, geometry, number theory, and more.

    ```bash theme={null}
    bench eval math
    ```
  </Card>

  <Card title="MATH-500" icon="fire">
    A challenging 500-problem subset of the MATH dataset.

    ```bash theme={null}
    bench eval math_500
    ```
  </Card>

  <Card title="MGSM" icon="globe">
    Multilingual Grade School Math - elementary math problems in multiple languages.

    ```bash theme={null}
    bench eval mgsm
    ```
  </Card>

  <Card title="MGSM English" icon="language">
    English-only version of Multilingual Grade School Math.

    ```bash theme={null}
    bench eval mgsm_en
    ```
  </Card>

  <Card title="MGSM Latin Scripts" icon="a">
    MGSM covering 5 languages using Latin scripts.

    ```bash theme={null}
    bench eval mgsm_latin
    ```
  </Card>

  <Card title="MGSM Non-Latin Scripts" icon="symbols">
    MGSM covering 6 languages using non-Latin scripts.

    ```bash theme={null}
    bench eval mgsm_non_latin
    ```
  </Card>

  <Card title="OTIS Mock AIME 2024" icon="pencil">
    Mock AIME problems from OTIS 2024.

    ```bash theme={null}
    bench eval otis_mock_aime_2024
    ```
  </Card>

  <Card title="OTIS Mock AIME 2025" icon="pencil">
    Mock AIME problems from OTIS 2025.

    ```bash theme={null}
    bench eval otis_mock_aime_2025
    ```
  </Card>
</CardGroup>

<br />

## Related Resources

* [AIME Competition](https://maa.org/maa-invitational-competitions/)
* [HMMT Tournament](https://www.hmmt.org/)
* [MATH Dataset Paper](https://arxiv.org/abs/2103.03874)
* [MGSM Paper](https://arxiv.org/abs/2210.03057)
