math benchmark
ArXivMath
ArXivMath is a final-answer benchmark of research-level mathematics maintained by MathArena. Problems are extracted monthly from recent arXiv paper abstracts, then filtered through automated and manual checks to ensure they are self-contained, non-trivial, and verifiable. Because problems are drawn from active research, the benchmark is more realistic and more closely connected to mathematical research than contest or olympiad benchmarks.
Updated Aug 7, 2026
Published models2
Registry coverage2
MetricScore
EvidenceB
ArXivMath leaderboard
Sorted by the source-provided rank. Higher score is better according to the registry.
| 1 | AN | 72.2% | 100.0% | 2 | C | |
| 2 | TE | 52.2% | 0.0% | 2 | C |
Score distribution
Top published rows on the benchmark's original scale.
ArXivMath highlights
The top published results on this benchmark's own scale.
What is ArXivMath?
Definition and scoring fields from the benchmark registry.
ArXivMath is a final-answer benchmark of research-level mathematics maintained by MathArena. Problems are extracted monthly from recent arXiv paper abstracts, then filtered through automated and manual checks to ensure they are self-contained, non-trivial, and verifiable. Because problems are drawn from active research, the benchmark is more realistic and more closely connected to mathematical research than contest or olympiad benchmarks.
Scores are shown in ratio. The current registry marks this benchmark as not independently verified with evidence level B.
- Family
- ArXivMath
- Modality
- text
- Primary category
- math
- Score direction
- higher
- LLMBoard eligible
- No
- Evaluation key
- arxivmath|llm-stats-current
Source-native results are preserved. Eligibility for the overall LLMBoard score is a separate policy decision.
FAQ
Common questions about ArXivMath.
Which model scores highest on ArXivMath?
Claude Sonnet 5 is currently ranked first with 72.2%.
What does ArXivMath measure?
ArXivMath is a final-answer benchmark of research-level mathematics maintained by MathArena. Problems are extracted monthly from recent arXiv paper abstracts, then filtered through automated and manual checks to ensure they are self-contained, non-trivial, and verifiable. Because problems are drawn from active research, the benchmark is more realistic and more closely connected to mathematical research than contest or olympiad benchmarks.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
2 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.