MathArena - BRUMO 2025 — leaderboard
Metric: Accuracy (%). Source: matharena.ai. 45 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 3 Flash | 100 |
| 2 | Step 3.5 Flash | 100 |
| 3 | Human Expert | 100 |
| 4 | GLM-5 | 99.17 |
| 5 | DeepSeek V3.2 Speciale | 99.17 |
| 6 | Gemini 3 Pro (Preview) | 98.33 |
| 7 | Kimi K2.5 (Thinking) | 98.33 |
| 8 | GPT-5.2 (High) | 98.33 |
| 9 | Grok 4.1 Fast (Reasoning) | 97.5 |
| 10 | DeepSeek V3.2 (Thinking) | 96.67 |
| 11 | Grok 4 Fast (Reasoning) | 95.83 |
| 12 | O3 (High) | 95.83 |
| 13 | DeepSeek V3.2 Exp (Thinking) | 95.83 |
| 14 | Grok 4 | 95 |
| 15 | GLM-4.6 | 94.17 |
Interactive version: theaggregate.ai/benchmark?slug=matharena-brumo-2025 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.