OpenCompass Reasoning - Inductive - English (CompassBench 2405): leaderboard
OpenCompass CompassBench 2405 edition (file dated 2024-06-12): inductive 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 | GPT-4 Turbo | 63.8 | #417 |
| 2 | GPT-4o (2024-05-13) | 51.8 | #337 |
| 3 | Yi Large | 47.9 | #566 |
| 4 | Claude 3 Opus | 40.1 | #565 |
| 5 | Qwen 1.5 110B Chat | 34.8 | #693 |
| 6 | Llama 3 8B Instruct | 34 | #1116 |
| 7 | Yi 1.5 34B Chat | 33.8 | #798 |
| 8 | DeepSeek V2 Chat | 33.6 | #572 |
| 9 | Qwen 1.5 72B Chat | 33.5 | #782 |
| 10 | Mistral Large | 33 | #713 |
| 11 | internlm2-chat-7B | 33 | #1113 |
| 12 | Moonshot v1 8K | 32.5 | #622 |
| 13 | Yi-1.5-9B Chat | 30.3 | #1032 |
| 14 | Nanbeige2-16B-Chat | 30 | #971 |
| 15 | Mixtral 8x22B Instruct | 29.5 | #738 |
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-inductive-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-11-05.