OpenCompass Knowledge - Common Sense (CompassBench 2601): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 26 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | O3 (2025-04-16) (High) | 95.5 |
| 2 | Gemini 3 Pro (Preview) | 95 |
| 3 | Kimi K2.5 | 93.8 |
| 4 | Tencent HY 2.0 Think | 93 |
| 5 | GPT-5.2 (High) | 92.8 |
| 6 | Ling-1T | 92 |
| 7 | Qwen 3 Max (2026-01-23) | 90.8 |
| 8 | Claude Opus 4.5 (Thinking) | 90.5 |
| 9 | GLM-4.7 | 90 |
| 10 | DeepSeek V3.2 | 89.5 |
| 11 | Claude Sonnet 4.5 (Thinking) | 89.5 |
| 12 | ERNIE 5.0 | 88.5 |
| 13 | Grok 4.1 | 88.2 |
| 14 | Ring-1T | 88 |
| 15 | Qwen 3 235B A22B 2507 (Thinking) | 87.8 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-knowledge-common-sense-compassbench-2601 · How It Works · Data refreshed daily, snapshot 2026-09-19.