OpenCompass Language - Dialogue - Chinese (CompassBench 2407): leaderboard
OpenCompass CompassBench 2407 edition (file dated 2024-08-05): dialogue 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. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4o (2024-05-13) | 65.9 | #337 |
| 2 | Mistral Large 2 (Jul) | 58.2 | #490 |
| 3 | Mistral Large | 58.2 | #713 |
| 4 | Nanbeige2-16B-Chat | 58.2 | #971 |
| 5 | Qwen 1.5 110B Chat | 55.8 | #693 |
| 6 | Yi 1.5 34B Chat | 54.8 | #798 |
| 7 | Yi-1.5-9B Chat | 53.9 | #1032 |
| 8 | GPT-4o Mini (2024-07-18) | 52.9 | #568 |
| 9 | Qwen 2 72B Instruct | 52.6 | #579 |
| 10 | InternLM2.5-Chat-7B | 52.4 | #1009 |
| 11 | Qwen 2 7B Instruct | 51.7 | #982 |
| 12 | Yi-1.5-6B-Chat | 51.2 | #1241 |
| 13 | GLM-4 9B Chat | 51 | #911 |
| 14 | Yi Large | 49 | #566 |
| 15 | GLM-4 (0520) | 48.3 | #612 |
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-dialogue-chinese-compassbench-2407 · How It Works · Data refreshed daily, snapshot 2026-11-05.