Open Arabic LLM - Arabic MMLU Islamic Studies (Primary School): leaderboard
The "Arabic MMLU Islamic Studies (Primary School)" subtask of Open Arabic LLM. A task from the Open Arabic LLM Leaderboard, evaluating models on Arabic-language knowledge, reasoning and understanding across native and translated benchmarks.
Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 163 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Yehia-7B-preview | 84.88 | #646 |
| 2 | ALLaM-7B-Instruct-preview | 84.78 | #718 |
| 3 | Qwen 2.5 72B Instruct | 84.08 | #454 |
| 4 | Saka-14B | 83.28 | #344 |
| 5 | gemma-3-27B-pt | 83.18 | #502 |
| 6 | AceGPT-v2-32B-Chat | 81.48 | #507 |
| 7 | Qwentile2.5-32B-Instruct | 81.28 | #339 |
| 8 | Awqward2.5-32B-Instruct | 81.18 | #371 |
| 9 | Qwen 2.5 14B | 80.98 | #725 |
| 10 | Qwen 2.5 32B Instruct | 80.88 | #498 |
| 11 | Qwen2.5-32B-Instruct-CFT | 80.88 | #375 |
| 12 | AIC-1 | 80.78 | #231 |
| 13 | Llama 3.3 70B Instruct | 80.68 | #534 |
| 14 | Qwen2.5-32B-Instruct-abliterated-v2 | 80.68 | #394 |
| 15 | Gemma 3 27B (IT) | 80.28 | #523 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-arabic-mmlu-islamic-studies-primary-school · How It Works · Data refreshed daily, snapshot 2026-10-20.