OpenCompass Language - Content Summarization - English (CompassBench 2405): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 39 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Command-R+ | 42.7 |
| 2 | internlm2-chat-20B | 38.2 |
| 3 | internlm2-chat-7B | 37.8 |
| 4 | chatglm3-6B | 35.6 |
| 5 | Llama 3 70B Instruct | 32.2 |
| 6 | Mixtral 8x22B Instruct (v0.1) | 30.6 |
| 7 | Mixtral 8x7B Instruct (v0.1) | 30.4 |
| 8 | DBRX | 30 |
| 9 | Llama 3 8B Instruct | 29.7 |
| 10 | Qwen 1.5 72B Chat | 29.5 |
| 11 | Qwen 1.5 110B Chat | 29 |
| 12 | Qwen 1.5 7B Chat | 28.8 |
| 13 | Qwen 1.5 14B Chat | 28.5 |
| 14 | GPT-3.5 Turbo | 26.4 |
| 15 | Mistral Large | 25.7 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-language-content-summarization-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-09-19.