OpenCompass Language - Dialogue - English (CompassBench 2407): leaderboard
OpenCompass CompassBench 2407 edition (file dated 2024-08-05): dialogue sub-task of the language 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. 36 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Qwen 1.5 110B Chat | 57.6 | #693 |
| 2 | Yi 1.5 34B Chat | 57.5 | #798 |
| 3 | GLM-4 9B Chat | 55.5 | #911 |
| 4 | Yi Large | 53.4 | #566 |
| 5 | Nanbeige2-16B-Chat | 52.2 | #971 |
| 6 | GPT-4o (2024-05-13) | 51.9 | #337 |
| 7 | Yi-1.5-9B Chat | 51.9 | #1032 |
| 8 | GLM-4 (0520) | 51.7 | #612 |
| 9 | GPT-4o Mini (2024-07-18) | 51.4 | #568 |
| 10 | Moonshot v1 8K | 49.5 | #622 |
| 11 | Llama 3.1 8B Instruct | 49 | #1017 |
| 12 | Llama 3.1 70B Instruct | 48.8 | #567 |
| 13 | Mistral Large 2 (Jul) | 47.1 | #490 |
| 14 | Mistral Large | 47.1 | #713 |
| 15 | Yi-1.5-6B-Chat | 47.1 | #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-dialogue-english-compassbench-2407 · How It Works · Data refreshed daily, snapshot 2026-11-05.