OpenCompass LLM - Language (CompassBench 2405): leaderboard
OpenCompass CompassBench 2405 edition (file dated 2024-06-12): language category score. 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 32B Chat | 60.9 | #820 |
| 2 | Qwen 1.5 110B Chat | 59.8 | #693 |
| 3 | Qwen 1.5 72B Chat | 59.6 | #782 |
| 4 | GPT-4o (2024-05-13) | 59.4 | #337 |
| 5 | GLM-4 | 57.7 | #647 |
| 6 | Qwen 1.5 14B Chat | 57 | #976 |
| 7 | Yi 1.5 34B Chat | 56.8 | #798 |
| 8 | Llama 3 70B Instruct | 55.8 | #640 |
| 9 | Yi Large | 55.7 | #566 |
| 10 | GPT-4 Turbo | 55.1 | #417 |
| 11 | Command-R+ | 53.1 | #869 |
| 12 | Yi-1.5-9B Chat | 52 | #1032 |
| 13 | internlm2-chat-20B | 51.4 | #988 |
| 14 | internlm2-chat-7B | 51.2 | #1113 |
| 15 | DeepSeek V2 Chat | 50 | #572 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=opencompass-llm-language-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-11-05.