math benchmark
Beyond AIME
Beyond AIME is a difficult mathematical reasoning benchmark designed to test deeper reasoning chains and harder decomposition than standard AIME-style problem sets.
Updated Aug 7, 2026
Published models5
Registry coverage5
MetricScore
EvidenceB
Beyond AIME leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | BY | 88.0% | 100.0% | 5 | C | |
| 2 | BY | 87.0% | 75.0% | 5 | C | |
| 3 | SA | 69.1% | 50.0% | 5 | C | |
| 4 | MA | 66.9% | 25.0% | 5 | C | |
| 5 | SA | 58.3% | 0.0% | 5 | C |
Score distribution
Top published rows on the benchmark's original scale.
Beyond AIME highlights
The top published results on this benchmark's own scale.
What is Beyond AIME?
Definition and scoring fields from the benchmark registry.
Beyond AIME is a difficult mathematical reasoning benchmark designed to test deeper reasoning chains and harder decomposition than standard AIME-style problem sets.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- Beyond AIME
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- beyond-aime|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about Beyond AIME.
Which model scores highest on Beyond AIME?
Seed 2.1 Turbo is currently ranked first with 88.0%.
What does Beyond AIME measure?
Beyond AIME is a difficult mathematical reasoning benchmark designed to test deeper reasoning chains and harder decomposition than standard AIME-style problem sets.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
5 unique published model results are currently shown.
Does this benchmark affect the overall score?
This benchmark is preserved as source-native evidence but is not eligible for the current overall score.