Skip to main content
Datasetbenchmarksv1.0

MATH Dataset

by UC Berkeley · open-source · Last verified 2026-03-17

A challenging benchmark of 12,500 competition mathematics problems from AMC, AIME, and similar competitions across 5 difficulty levels and 7 subjects. Each problem includes a full step-by-step solution in LaTeX, making it suitable for both evaluation and training of mathematical reasoning.

https://huggingface.co/datasets/hendrycks/competition_math
B+
B+Good
Adoption: AQuality: A+Freshness: B+Citations: A+Engagement: F

Specifications

License
MIT
Pricing
open-source
Capabilities
math-evaluation, advanced-reasoning-benchmark, step-by-step-solutions
Integrations
huggingface-datasets, lm-eval-harness
Use Cases
model-evaluation, advanced-math-reasoning, mathematical-training
API Available
No
Tags
benchmark, competition-math, hard-math, step-by-step, latex
Added
2026-03-17
Completeness
100%

Index Score

77.3
Adoption
88
Quality
93
Freshness
72
Citations
94
Engagement
0

Put AI to work for your business

Deploy this dataset alongside autonomous AaaS agents that handle tasks end-to-end — no babysitting required.

Explore the full AI ecosystem on Agents as a Service