OpenCompass Knowledge - Social Science - Chinese (CompassBench 2504): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 28 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4.1 (2025-04-14) | 92 |
| 2 | GPT-4.1 Mini | 91 |
| 3 | Grok 3 | 91 |
| 4 | Llama 4 Maverick Instruct | 91 |
| 5 | Gemini 2.5 Pro (Preview 05-06) | 91 |
| 6 | O4 Mini (High) | 89 |
| 7 | Step 2 16K | 89 |
| 8 | DeepSeek V3 (0324) | 88 |
| 9 | O3 Mini (High) | 87 |
| 10 | DeepSeek R1 | 86 |
| 11 | ERNIE-X1-Turbo-32K | 86 |
| 12 | Claude 3.7 Sonnet (Thinking) | 85 |
| 13 | Gemma 3 27B (IT) | 84 |
| 14 | Qwen 3 235B A22B (Thinking) | 84 |
| 15 | Llama 4 Scout Instruct | 83 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-knowledge-social-science-chinese-compassbench-2504 · How It Works · Data refreshed daily, snapshot 2026-09-19.