OpenCompass LLM - Average - English (CompassBench 2402): leaderboard
OpenCompass CompassBench 2402 edition (file dated 2024-02-20): overall average across the edition's capability categories, 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. 35 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | GPT-4 Turbo (Preview) | 64.6 | #416 |
| 2 | GLM-4 | 59.7 | #647 |
| 3 | Qwen Max | 56.3 | #388 |
| 4 | MiniMax-abab5.5 | 53.7 | #1030 |
| 5 | internlm2-chat-20B | 52.8 | #988 |
| 6 | GPT-3.5 Turbo | 52.5 | #858 |
| 7 | Qwen 1.5 72B Chat | 52.4 | #782 |
| 8 | Qwen-72B-Chat | 52.1 | #835 |
| 9 | internlm2-chat-7B | 49.2 | #1113 |
| 10 | Qwen 1.5 14B Chat | 49 | #976 |
| 11 | Yi Chat 34B | 47.9 | #950 |
| 12 | OrionStar-Yi-34B-Chat | 46 | #872 |
| 13 | Qwen-14B-Chat | 45.2 | #1161 |
| 14 | Mixtral 8x7B Instruct | 43.9 | #1085 |
| 15 | Llama 2 70B Chat | 40.8 | #1275 |
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-average-english-compassbench-2402 · How It Works · Data refreshed daily, snapshot 2026-11-05.