LLM Stats (ArXivMath): leaderboard
MathArena's ArXivMath as listed by LLM Stats: final-answer research mathematics problems extracted monthly from recent arXiv abstracts and checked to be self-contained and verifiable. Accuracy in percent; higher is better.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around February 2027. 5 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 5.5 | 91.2 | #5 |
| 2 | Claude Sonnet 5.5 | 86.8 | #19 |
| 3 | Hy4 preview | 66.6 | #68 |
| 4 | Hy3 | 52.2 | #195 |
No result here: #1 Claude Fable 5.2, #2 Gemini 4 Carbon, #6 GPT-6 Astra.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-arxivmath · How It Works · Data refreshed daily, snapshot 2026-11-13.