OpenCompass Language - NLP - English (CompassBench 2411): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 34 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Grok Beta | 63.4 |
| 2 | GPT-4o (2024-11-20) | 62.8 |
| 3 | Yi Lightning | 61.9 |
| 4 | O1 Mini (2024-09-12) | 61.2 |
| 5 | Qwen 2.5 72B Instruct | 60.9 |
| 6 | GLM-4 Plus | 58.7 |
| 7 | Claude 3.5 Sonnet (20241022) | 58.5 |
| 8 | Step 2 16K | 58.3 |
| 9 | TeleChat2 | 58 |
| 10 | Mistral Large 2 (Nov) Instruct (2411) | 56.2 |
| 11 | Llama 3.1 405B Instruct FP8 | 56.2 |
| 12 | DeepSeek V2.5 | 54.4 |
| 13 | Gemma 2 27B (IT) | 53 |
| 14 | Llama 3.1 70B Instruct | 51.5 |
| 15 | Yi 1.5 34B Chat | 51 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-language-nlp-english-compassbench-2411 · How It Works · Data refreshed daily, snapshot 2026-09-19.