OpenCompass Knowledge - Common Sense - Chinese (CompassBench 2405): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 39 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 1.5 72B Chat | 78.7 |
| 2 | GPT-4o (2024-05-13) | 77 |
| 3 | Qwen 1.5 110B Chat | 76.5 |
| 4 | Yi 1.5 34B Chat | 76 |
| 5 | Qwen 1.5 32B Chat | 75.4 |
| 6 | Yi Large | 74.9 |
| 7 | GLM-4 | 74.9 |
| 8 | Qwen 1.5 14B Chat | 73.2 |
| 9 | GPT-4 Turbo | 72.1 |
| 10 | Moonshot v1 8K | 66.1 |
| 11 | Yi-1.5-9B Chat | 66.1 |
| 12 | DeepSeek V2 Chat | 65 |
| 13 | Llama 3 70B Instruct | 64.5 |
| 14 | Claude 3 Opus | 63.9 |
| 15 | internlm2-chat-20B | 59.6 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-knowledge-common-sense-chinese-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-09-19.