MathArena - HMMT Feb 2025: leaderboard
Harvard-MIT Mathematics Tournament February 2025 problems, a prestigious undergraduate-organized competition spanning individual and team rounds across multiple math topics.
Metric: Accuracy (%). Source: matharena.ai. Source: benchmarklist.com. Saturation forecast: Estimated already saturated. 60 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-5.2 (xHigh) | 100 | #110 (GPT-5.2) |
| 2 | GPT-5.2 (High) | 98.33 | #110 (GPT-5.2) |
| 3 | Step 3.5 Flash | 98.33 | #270 |
| 4 | Gemini 3 Flash | 97.5 | #99 |
| 5 | Gemini 3 Pro (Preview) | 97.5 | #66 |
| 6 | GLM-5 | 97.5 | #141 |
| 7 | DeepSeek V3.2 Speciale | 97.5 | #128 |
| 8 | GPT-5.5 | 96.5 | #31 |
| 9 | Claude Opus 4.8 | 96.5 | #37 |
| 10 | Grok 4 | 95 | #181 |
| 11 | Qwen 3.7 Max | 95 | #75 |
| 12 | Gemini 3.1 Pro (Preview) | 94.8 | #58 |
| 13 | DeepSeek V4 Pro | 94.4 | #102 |
| 14 | GLM-5.2 | 94.4 | #73 |
| 15 | GLM-5.1 | 94 | #97 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=matharena-hmmt-feb-2025 · How It Works · Data refreshed daily, snapshot 2026-10-20.