LLM Stats (Beyond AIME): leaderboard
LLM Stats mirror of Beyond AIME, evaluating advanced competition-math reasoning beyond standard AIME-style problems.
Metric: Score (%). Source: llm-stats.com. Saturation forecast: Around January 2027. 7 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Seed 2.1 Turbo | 88 | #86 |
| 2 | Seed 2.1 Pro | 87 | #67 |
| 3 | Seed 1.8 | 77 | #139 |
| 4 | Sarvam 105B | 69.1 | #409 |
| 5 | Seed 2.0 Mini | 69 | #276 |
| 6 | Mistral Medium 3.5 | 66.9 | #353 |
| 7 | Sarvam 30B | 58.3 | #719 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=llm-stats-beyond-aime · How It Works · Data refreshed daily, snapshot 2026-10-20.