OpenCompass Language - Content Summarization - Chinese (CompassBench 2405): leaderboard
OpenCompass CompassBench 2405 edition (file dated 2024-06-12): content summarization 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 | internlm2-chat-20B | 33.2 | #988 |
| 2 | internlm2-chat-7B | 31 | #1113 |
| 3 | Command-R+ | 27.6 | #869 |
| 4 | Mixtral 8x22B Instruct | 26.3 | #738 |
| 5 | Qwen 1.5 14B Chat | 25.9 | #976 |
| 6 | Qwen 1.5 110B Chat | 25.7 | #693 |
| 7 | chatglm3-6B | 25.4 | #1416 |
| 8 | DeepSeek-V2-Lite-Chat | 24.1 | #1189 |
| 9 | Qwen 1.5 72B Chat | 23.8 | #782 |
| 10 | GPT-3.5 Turbo | 23.7 | #858 |
| 11 | Qwen 1.5 32B Chat | 22.2 | #820 |
| 12 | GPT-4 Turbo | 22 | #417 |
| 13 | Yi 1.5 34B Chat | 21.3 | #798 |
| 14 | Qwen 1.5 7B Chat | 21.3 | #1292 |
| 15 | Yi-1.5-6B-Chat | 20.8 | #1241 |
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-content-summarization-chinese-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-11-05.