OpenCompass Reasoning - Common Sense - English (CompassBench 2405): leaderboard
OpenCompass CompassBench 2405 edition (file dated 2024-06-12): common-sense reasoning sub-task of the reasoning category, English questions. CompassBench is OpenCompass's periodic evaluation of chat models on a private question set; each edition is a separate board. Score 0-100, higher is better.
Metric: Score (%). Source: rank.opencompass.org.cn. Saturation forecast: Estimated already saturated. 39 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Llama 3 70B Instruct | 31.8 | #640 |
| 2 | Yi 1.5 34B Chat | 30 | #798 |
| 3 | GPT-4o (2024-05-13) | 26.9 | #337 |
| 4 | Yi Large | 26.5 | #566 |
| 5 | Yi-1.5-9B Chat | 26 | #1032 |
| 6 | Claude 3 Opus | 25.1 | #565 |
| 7 | Qwen 1.5 32B Chat | 25.1 | #820 |
| 8 | Qwen 1.5 110B Chat | 25.1 | #693 |
| 9 | internlm2-chat-7B | 24.2 | #1113 |
| 10 | Command-R+ | 23.8 | #869 |
| 11 | Llama 3 8B Instruct | 22.9 | #1116 |
| 12 | GLM-4 | 20.6 | #647 |
| 13 | DBRX | 20.2 | #1041 |
| 14 | Mistral 7B Instruct (v0.3) | 19.7 | #1255 |
| 15 | Mixtral 8x7B Instruct | 19.3 | #1085 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=opencompass-reasoning-common-sense-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-11-05.