OpenCompass Language - Creation - English (CompassBench 2411): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 34 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | GPT-4o (2024-11-20) | 59.3 |
| 2 | O1 Mini (2024-09-12) | 57.2 |
| 3 | GLM-4 Plus | 54.8 |
| 4 | Yi Lightning | 54.5 |
| 5 | Grok Beta | 53.8 |
| 6 | Qwen 2.5 72B Instruct | 52 |
| 7 | Mistral Large 2 (Nov) Instruct (2411) | 50.8 |
| 8 | TeleChat2 | 50.3 |
| 9 | DeepSeek V2.5 | 50.1 |
| 10 | GPT-4o (2024-08-06) | 50 |
| 11 | Claude 3.5 Sonnet (20241022) | 49.3 |
| 12 | Yi 1.5 34B Chat | 48.4 |
| 13 | 360gpt2-pro | 48.2 |
| 14 | Mistral-Small-Instruct-2409 | 48.1 |
| 15 | Step 2 16K | 46.9 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-language-creation-english-compassbench-2411 · How It Works · Data refreshed daily, snapshot 2026-09-19.