math benchmark
FrontierMath Leaderboard
FrontierMath is a benchmark of hundreds of original, exceptionally challenging mathematics problems crafted and vetted by expert mathematicians, covering major branches of modern mathematics from number theory and real analysis to algebraic geometry and category theory.
Updated Sep 5, 2026
Models17
Model coverage17
MetricScore
EvidenceB
FrontierMath Ranking
Higher score ranks better on this benchmark.
Sort by
Rank | Model | Score | Percentile | Participants | Evidence | Evaluated |
|---|
| Rank01 | ModelOP | Score89.00% | Percentile100.00% | Participants17 | EvidenceC | Evaluated |
| Rank02 | ModelOP | Score84.90% | Percentile93.75% | Participants17 | EvidenceC | Evaluated |
| Rank03 | ModelOP | Score78.60% | Percentile87.50% | Participants17 | EvidenceC | Evaluated |
| Rank04 | ModelOP | Score47.60% | Percentile81.25% | Participants17 | EvidenceC | Evaluated |
| Rank05 | ModelOP | Score40.30% | Percentile75.00% | Participants17 | EvidenceC | Evaluated |
| Rank06 | ModelOP | Score39.60% | Percentile68.75% | Participants17 | EvidenceC | Evaluated |
| Rank07 | ModelOP | Score35.40% | Percentile62.50% | Participants17 | EvidenceC | Evaluated |
| Rank08 | ModelOP | Score26.70% | Percentile56.25% | Participants17 | EvidenceC | Evaluated |
| Rank09 | ModelOP | Score26.70% | Percentile50.00% | Participants17 | EvidenceC | Evaluated |
| Rank10 | ModelOP | Score26.70% | Percentile43.75% | Participants17 | EvidenceC | Evaluated |
| Rank11 | ModelOP | Score26.30% | Percentile37.50% | Participants17 | EvidenceC | Evaluated |
| Rank12 | ModelOP | Score22.10% | Percentile31.25% | Participants17 | EvidenceC | Evaluated |
| Rank13 | ModelOP | Score15.80% | Percentile25.00% | Participants17 | EvidenceC | Evaluated |
| Rank14 | ModelOP | Score9.60% | Percentile18.75% | Participants17 | EvidenceC | Evaluated |
| Rank15 | ModelOP | Score9.20% | Percentile12.50% | Participants17 | EvidenceC | Evaluated |
| Rank16 | ModelMI | Score6.30% | Percentile6.25% | Participants17 | EvidenceC | Evaluated |
| Rank17 | ModelOP | Score5.50% | Percentile0.00% | Participants17 | EvidenceC | Evaluated |
FrontierMath Highlights
The leading models and scores on this benchmark.
FrontierMath Score Distribution
A closer view of the leading scores on this benchmark.
The Top AI Models for FrontierMath
The first five results on this benchmark, with official price and output speed added where the model identity can be matched.
What is FrontierMath?
What FrontierMath measures and how its scores work.
FrontierMath is a mathematics benchmark consisting of hundreds of original problems crafted and vetted by expert mathematicians.
It reports a Score as a ratio.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- FrontierMath
- 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 FrontierMath.
Which model scores highest on FrontierMath?
GPT-5.6 Sol is currently ranked first with 89.00%.
What are the top three models on FrontierMath?
The current leaders are GPT-5.6 Sol (89.00%), GPT-5.6 Terra (84.90%), and GPT-5.6 Luna (78.60%).
Which FrontierMath model has the lowest official input price?
GPT-5 nano has the lowest matched official input price at $0.05 input / $0.40 output per 1M tokens.
Which models are fastest among FrontierMath results?
The fastest matched records are GPT-5 nano (500.00 tok/s via OpenAI), GPT-5 mini (167.61 tok/s via OpenAI), and GPT-5.5 (134.94 tok/s via OpenAI).
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.