Open LLM Leaderboard - MATH Level 5: leaderboard
The "MATH Level 5" subtask of Open LLM Leaderboard. A task from the Open LLM Leaderboard, the standard open-weight evaluation suite spanning knowledge, reasoning, math and truthfulness benchmarks.
Metric: Score. Source: huggingface.co. Saturation forecast: Estimated already saturated. 4576 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 2.5 32B Instruct | 62.54 | #498 |
| 2 | Qwen2.5-Math-72B-Instruct | 62.39 | #717 |
| 3 | Awqward2.5-32B-Instruct | 62.31 | #371 |
| 4 | Qwen 2.5 72B Instruct | 59.82 | #454 |
| 5 | Qwen2.5-32B-Instruct-abliterated-v2 | 59.52 | #394 |
| 6 | Qwen2.5-Math-7B-Instruct | 58.08 | #1444 |
| 7 | DeepSeek R1 Distill Qwen 14B | 57.02 | #844 |
| 8 | Qwen2.5-Math-7B-CFT | 55.74 | #1438 |
| 9 | tempmotacilla-cinerea-0308 | 55.51 | #395 |
| 10 | Qwen 2.5 14B Instruct | 55.29 | #647 |
| 11 | lambda-qwen2.5-14B-dpo-test | 54.61 | #481 |
| 12 | Lamarckvergence-14B | 54 | #545 |
| 13 | Qwen2.5-Lumen-14B | 53.63 | #477 |
| 14 | Qwen2.5-14B-Gutenberg-Instruct-Slerpeno | 53.25 | #442 |
| 15 | Qwen2.5-14B-Instruct-1M | 53.02 | #578 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-llm-leaderboard-math-level-5 · How It Works · Data refreshed daily, snapshot 2026-10-20.