Arcee AIME 2025 (BenchLM): leaderboard
Arcee’s launch-chart comparison on the AIME 2025 integer-answer mathematics examination. This source-specific snapshot is kept distinct from other AIME evaluation protocols. These are published results, not BenchLM ability estimates.
Metric: Accuracy (%). Source: benchlm.ai. Saturation forecast: Estimated already saturated. 6 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Claude Opus 4.6 | 99.8 | #67 |
| 2 | Kimi K2.5 | 96.3 | #152 |
| 3 | Trinity-Large-Thinking | 96.3 | #371 |
| 4 | GLM-5 | 93.3 | #149 |
| 5 | MiniMax-M2.7 | 80 | #270 |
| 6 | trinity-large-preview | 24 | #591 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=arcee-aime-2025-benchlm · How It Works · Data refreshed daily, snapshot 2026-11-05.