OpenCompass Reasoning - Humanities - English (CompassBench 2409): leaderboard
Metric: Score (%). Source: rank.opencompass.org.cn. 30 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 63.8 |
| 2 | Step 2 16K | 62.3 |
| 3 | Claude 3.5 Sonnet (20240620) | 61 |
| 4 | DeepSeek V2.5 | 61 |
| 5 | GLM-4 Plus | 58 |
| 6 | GPT-4o (2024-05-13) | 56.2 |
| 7 | Mistral Large 2 (Jul) | 51.9 |
| 8 | Mistral Nemo Instruct (2407) | 50.2 |
| 9 | Llama 3.1 70B Instruct | 50 |
| 10 | Llama 3.1 405B Instruct FP8 | 49.5 |
| 11 | GPT-4o Mini (2024-07-18) | 49.3 |
| 12 | GPT-4o (2024-08-06) | 48.6 |
| 13 | Mistral-Small-Instruct-2409 | 47.5 |
| 14 | Gemini 1.5 Pro | 47.2 |
| 15 | Qwen 2.5 7B Instruct | 45.9 |
Interactive version: theaggregate.ai/benchmark?slug=opencompass-reasoning-humanities-english-compassbench-2409 · How It Works · Data refreshed daily, snapshot 2026-09-19.