OpenCompass Math - College - Chinese (CompassBench 2409): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 30 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 71.5 |
| 2 | Mistral Large 2 (Jul) | 70.5 |
| 3 | GLM-4 Plus | 70 |
| 4 | Claude 3.5 Sonnet (20240620) | 65 |
| 5 | GPT-4o (2024-05-13) | 65 |
| 6 | GPT-4o (2024-08-06) | 63 |
| 7 | Qwen 2.5 7B Instruct | 63 |
| 8 | Llama 3.1 405B Instruct FP8 | 63 |
| 9 | DeepSeek V2.5 | 62 |
| 10 | Step 2 16K | 60 |
| 11 | Gemini 1.5 Pro | 59.5 |
| 12 | GPT-4o Mini (2024-07-18) | 58 |
| 13 | GLM-4 9B Chat | 53 |
| 14 | Mistral-Small-Instruct-2409 | 52.5 |
| 15 | Yi Large | 52.5 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-math-college-chinese-compassbench-2409 · How It Works · Data refreshed daily, snapshot 2026-09-19.