---
viewer: false
tags:
- benchpress
- dataset-composition
---

# math-reasoning-mix

Fixed evaluation-set composition built with BenchPress. This repo stores
only the recipe (`manifest.json`) — source data is streamed at load time.

**Abilities:** quantitative_manipulation, algorithmic_decomposition

## Sources

| benchmark | source repo | config | split | n_samples | license | gated |
|---|---|---|---|---|---|---|
| GSM8K | [openai/gsm8k](https://huggingface.co/datasets/openai/gsm8k) | main | test | 200 | mit | no |
| MATH-500 | [HuggingFaceH4/MATH-500](https://huggingface.co/datasets/HuggingFaceH4/MATH-500) | - | test | 150 | unknown | no |
| AIME 2024 | [math-ai/aime24](https://huggingface.co/datasets/math-ai/aime24) | - | test | 30 | apache-2.0 | no |

## Reference scores (expected on this composition)

| model | expected score |
|---|---|
| Claude-3.5-Sonnet | 0.8249 |
| Claude-Sonnet-4 | 0.909 |
| DeepSeek-v3 | 0.857 |
| GLM-4.7 | 0.9337 |
| GPT-4o | 0.8211 |
| GPT-oss-120b | 0.9639 |
| GPT-oss-20b | 0.9271 |
| Gemma-3-4B | 0.3261 |
| KIMI-K2 | 0.9394 |
| Llama-3.1-8B | 0.6016 |
| Phi-4-Mini | 0.7576 |
| Qwen2.5-72B | 0.8491 |
| Qwen3-235B | 0.9513 |

## Usage

```python
from benchpress_hub import load_composition

ds = load_composition("seonghyeon0408/math-reasoning-mix")
```

Sources pinned to commit SHAs; loading is deterministic (fixed items, seeded order).