OpenCompass Knowledge - Humanities - Chinese (CompassBench 2501): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 40 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Gemini 2.0 Pro (Preview 02-05) | 96 |
| 2 | O1 (2024-12-17) | 92 |
| 3 | Mistral Large 2 (Nov) Instruct (2411) | 92 |
| 4 | Qwen 2.5 Max | 91 |
| 5 | Claude 3.5 Sonnet (20241022) | 89 |
| 6 | GLM-4 Plus | 88 |
| 7 | GPT-4o (2024-11-20) | 87 |
| 8 | DeepSeek V3 | 86 |
| 9 | Llama 3.1 405B Instruct FP8 | 86 |
| 10 | Yi Lightning | 86 |
| 11 | DeepSeek R1 | 85 |
| 12 | Gemini 2.0 Flash (001) | 85 |
| 13 | Qwen 2.5 72B Instruct | 84 |
| 14 | Doubao-1.5-pro-32k-250115 | 82 |
| 15 | GPT-4o Mini (2024-07-18) | 81 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-knowledge-humanities-chinese-compassbench-2501 · How It Works · Data refreshed daily, snapshot 2026-09-19.