OpenCompass LLM - Math (CompassBench 2409): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 30 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 77 |
| 2 | Mistral Large 2 (Jul) | 73.7 |
| 3 | Claude 3.5 Sonnet (20240620) | 72.1 |
| 4 | GPT-4o (2024-05-13) | 70.6 |
| 5 | GPT-4o (2024-08-06) | 70.5 |
| 6 | Llama 3.1 405B Instruct FP8 | 70.1 |
| 7 | GLM-4 Plus | 67.6 |
| 8 | Qwen 2.5 7B Instruct | 66.9 |
| 9 | DeepSeek V2.5 | 66.5 |
| 10 | Step 2 16K | 62.1 |
| 11 | Gemini 1.5 Pro | 61.3 |
| 12 | Llama 3.1 70B Instruct | 58.7 |
| 13 | Yi Large | 55.5 |
| 14 | GPT-4o Mini (2024-07-18) | 54.5 |
| 15 | Mistral-Small-Instruct-2409 | 51.6 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-math-compassbench-2409 · How It Works · Data refreshed daily, snapshot 2026-09-19.