OpenCompass LLM - Knowledge - English (CompassBench 2405): leaderboard
OpenCompass CompassBench 2405 edition (file dated 2024-06-12): knowledge category score, 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. 39 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Mixtral 8x22B Instruct | 83.7 | #738 |
| 2 | Claude 3 Opus | 83.5 | #565 |
| 3 | Mistral Large | 82.9 | #713 |
| 4 | DBRX | 81.9 | #1041 |
| 5 | Llama 3 70B Instruct | 81.7 | #640 |
| 6 | Command-R+ | 81.2 | #869 |
| 7 | GPT-4 Turbo | 80.9 | #417 |
| 8 | Yi Large | 80.1 | #566 |
| 9 | Qwen 1.5 110B Chat | 79.7 | #693 |
| 10 | DeepSeek V2 Chat | 79.4 | #572 |
| 11 | Qwen 1.5 72B Chat | 76.8 | #782 |
| 12 | Mixtral 8x7B Instruct | 76.5 | #1085 |
| 13 | Moonshot v1 8K | 76 | #622 |
| 14 | GPT-4o (2024-05-13) | 75.4 | #337 |
| 15 | GLM-4 | 75 | #647 |
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-knowledge-english-compassbench-2405 · How It Works · Data refreshed daily, snapshot 2026-11-05.