OpenCompass Reasoning - Deductive - English (CompassBench 2405): leaderboard
OpenCompass CompassBench 2405 edition (file dated 2024-06-12): deductive 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 | Qwen 1.5 110B Chat | 23.3 | #693 |
| 2 | Yi Large | 23.3 | #566 |
| 3 | Yi-1.5-9B Chat | 20.9 | #1032 |
| 4 | Yi-1.5-6B-Chat | 20.9 | #1241 |
| 5 | internlm2-chat-7B | 20.9 | #1113 |
| 6 | Llama 3 70B Instruct | 18.6 | #640 |
| 7 | Qwen 1.5 72B Chat | 18.6 | #782 |
| 8 | Mistral 7B Instruct (v0.3) | 16.3 | #1255 |
| 9 | GPT-4 Turbo | 16.3 | #417 |
| 10 | Yi 1.5 34B Chat | 16.3 | #798 |
| 11 | DBRX | 16.3 | #1041 |
| 12 | Llama 3 8B Instruct | 13.9 | #1116 |
| 13 | GPT-4o (2024-05-13) | 13.9 | #337 |
| 14 | Command-R+ | 13.9 | #869 |
| 15 | Qwen 1.5 7B Chat | 13.9 | #1292 |
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-deductive-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-11-05.