Open Arabic LLM - Arabic MMLU General Knowledge: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 163 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen 2.5 72B Instruct | 70.83 |
| 2 | Qwen2.5-32B-Instruct-abliterated-v2 | 68.98 |
| 3 | Awqward2.5-32B-Instruct | 68.87 |
| 4 | Qwen 2.5 32B Instruct | 68.75 |
| 5 | gemma-3-27B-pt | 68.52 |
| 6 | Qwen2.5-32B-Instruct-CFT | 68.4 |
| 7 | Llama 3.3 70B Instruct | 68.17 |
| 8 | Gemma 3 27B (IT) | 67.48 |
| 9 | Saka-14B | 67.48 |
| 10 | EZO-Qwen2.5-32B-Instruct | 67.48 |
| 11 | Qwentile2.5-32B-Instruct | 66.9 |
| 12 | aya-expanse-32B | 65.28 |
| 13 | tempmotacilla-cinerea-0308 | 65.16 |
| 14 | gemma-3-12B-pt | 63.89 |
| 15 | Qwen 2.5 72B | 63.66 |
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-arabic-mmlu-general-knowledge · How It Works · Data refreshed daily, snapshot 2026-09-05.