OpenCompass Language - Intention Recognition - Chinese (CompassBench 2405): leaderboard
OpenCompass CompassBench 2405 edition (file dated 2024-06-12): intention recognition sub-task of the language category, Chinese 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 | 78.7 | #693 |
| 2 | Qwen 1.5 32B Chat | 76.3 | #820 |
| 3 | Qwen 1.5 72B Chat | 71 | #782 |
| 4 | internlm2-chat-7B | 70.3 | #1113 |
| 5 | Qwen 1.5 14B Chat | 69 | #976 |
| 6 | internlm2-chat-20B | 68.3 | #988 |
| 7 | GPT-4o (2024-05-13) | 61 | #337 |
| 8 | Yi 1.5 34B Chat | 61 | #798 |
| 9 | GLM-4 | 58 | #647 |
| 10 | Yi Large | 56.7 | #566 |
| 11 | Yi-1.5-9B Chat | 55.3 | #1032 |
| 12 | Llama 3 70B Instruct | 55 | #640 |
| 13 | Qwen 1.5 7B Chat | 52 | #1292 |
| 14 | Command-R+ | 40.3 | #869 |
| 15 | Moonshot v1 8K | 37.7 | #622 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=opencompass-language-intention-recognition-chinese-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-11-05.