Open Arabic LLM - Arabic MMLU HT Business Ethics: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 163 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instruct | 77 |
| 2 | Qwen 2.5 72B Instruct | 77 |
| 3 | Qwen 2 72B | 73 |
| 4 | gemma-3-27B-pt | 72 |
| 5 | Qwen 2.5 72B | 72 |
| 6 | Saka-14B | 72 |
| 7 | Gemma 2 27B (IT) | 71 |
| 8 | Awqward2.5-32B-Instruct | 71 |
| 9 | Qwen2.5-Lumen-14B | 71 |
| 10 | Qwentile2.5-32B-Instruct | 70 |
| 11 | free-evo-qwen72B-v0.8-re | 70 |
| 12 | Rombos-LLM-V2.6-Qwen-14B | 70 |
| 13 | Gemma 3 27B (IT) | 69 |
| 14 | Qwen 2.5 32B Instruct | 69 |
| 15 | Qwen 2.5 14B Instruct | 69 |
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-arabic-mmlu-ht-business-ethics · How It Works · Data refreshed daily, snapshot 2026-09-05.