math benchmark
MATH Leaderboard
MATH is a dataset of 12,500 challenging competition mathematics problems from AMC 10, AMC 12, AIME, and other mathematics competitions.
Updated Sep 6, 2026
Models71
Model coverage71
MetricScore
EvidenceB
MATH Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelOP | Score97.90% | Percentile100.00% | Participants71 | EvidenceC | Evaluated |
| Rank02 | ModelOP | Score96.40% | Percentile98.57% | Participants71 | EvidenceC | Evaluated |
| Rank03 | ModelMA | Score90.40% | Percentile97.14% | Participants71 | EvidenceC | Evaluated |
| Rank04 | ModelMA | Score90.40% | Percentile95.71% | Participants71 | EvidenceC | Evaluated |
| Rank05 | ModelGO | Score89.70% | Percentile94.29% | Participants71 | EvidenceC | Evaluated |
| Rank06 | ModelMA | Score89.10% | Percentile92.86% | Participants71 | EvidenceC | Evaluated |
| Rank07 | ModelGO | Score89.00% | Percentile91.43% | Participants71 | EvidenceC | Evaluated |
| Rank08 | ModelMA | Score87.60% | Percentile90.00% | Participants71 | EvidenceC | Evaluated |
| Rank09 | ModelGO | Score86.80% | Percentile88.57% | Participants71 | EvidenceC | Evaluated |
| Rank10 | ModelGO | Score86.50% | Percentile87.14% | Participants71 | EvidenceC | Evaluated |
| Rank11 | ModelXI | Score86.20% | Percentile85.71% | Participants71 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score85.50% | Percentile84.29% | Participants71 | EvidenceC | Evaluated |
| Rank13 | ModelOP | Score84.70% | Percentile82.86% | Participants71 | EvidenceC | Evaluated |
| Rank14 | ModelGO | Score83.80% | Percentile81.43% | Participants71 | EvidenceC | Evaluated |
| Rank15 | ModelAC | Score83.10% | Percentile80.00% | Participants71 | EvidenceC | Evaluated |
| Rank16 | ModelAC | Score83.10% | Percentile78.57% | Participants71 | EvidenceC | Evaluated |
| Rank17 | ModelMA | Score83.00% | Percentile77.14% | Participants71 | EvidenceC | Evaluated |
| Rank18 | ModelAC | Score82.20% | Percentile75.71% | Participants71 | EvidenceC | Evaluated |
| Rank19 | ModelMI | Score80.40% | Percentile74.29% | Participants71 | EvidenceC | Evaluated |
| Rank20 | ModelAC | Score80.00% | Percentile72.86% | Participants71 | EvidenceC | Evaluated |
| Rank21 | ModelAN | Score78.30% | Percentile71.43% | Participants71 | EvidenceC | Evaluated |
| Rank22 | ModelGO | Score77.90% | Percentile70.00% | Participants71 | EvidenceC | Evaluated |
| Rank23 | ModelME | Score77.00% | Percentile68.57% | Participants71 | EvidenceC | Evaluated |
| Rank24 | ModelOP | Score76.60% | Percentile67.14% | Participants71 | EvidenceC | Evaluated |
| Rank25 | ModelAM Nova ProAmazon | Score76.60% | Percentile65.71% | Participants71 | EvidenceC | Evaluated |
| Rank26 | ModelXA | Score76.10% | Percentile64.29% | Participants71 | EvidenceC | Evaluated |
| Rank27 | ModelGO | Score75.60% | Percentile62.86% | Participants71 | EvidenceC | Evaluated |
| Rank28 | ModelAC | Score75.50% | Percentile61.43% | Participants71 | EvidenceC | Evaluated |
| Rank29 | ModelDE | Score74.70% | Percentile60.00% | Participants71 | EvidenceC | Evaluated |
| Rank30 | ModelME | Score73.80% | Percentile58.57% | Participants71 | EvidenceC | Evaluated |
MATH Highlights
The leading models and scores on this benchmark.
MATH Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for MATH
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is MATH?
What MATH measures and how its scores work.
MATH is a mathematics benchmark dataset with problems spanning difficulty levels 1-5 and seven subjects: Prealgebra, Algebra, Number Theory, Counting and Probability, Geometry, Intermediate Algebra, and Precalculus.
MATH reports Score as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- MATH
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- overall
LLM Stats. Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about MATH.
Which model scores highest on MATH?
o3-mini is currently ranked first with 97.90%.
What are the top three models on MATH?
The current leaders are o3-mini (97.90%), o1 (96.40%), and MiniStral 3 (14B Instruct 2512) (90.40%).
Which MATH model has the lowest official input price?
Nova Micro has the lowest matched official input price at $0.04 input / $0.14 output per 1M tokens.
Which models are fastest among MATH results?
The fastest matched records are Llama 3.3 70B Instruct (2,220.00 tok/s via Cerebras), Llama 4 Scout (776.10 tok/s via Groq), and Llama 4 Maverick (307.30 tok/s via Groq).
Does the highest score result prove overall model quality?
No. This benchmark measures one defined capability or task. The overall LLMBoard score uses a separate aggregation across eligible benchmark evidence.