Open Arabic LLM - Arabic MMLU Arabic Language (General): leaderboard
The "Arabic MMLU Arabic Language (General)" 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: Around January 2027. 163 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Saka-14B | 67.65 | #344 |
| 2 | gemma-3-27B-pt | 67.48 | #502 |
| 3 | free-evo-qwen72B-v0.8-re | 65.85 | #435 |
| 4 | AceGPT-v2-32B-Chat | 65.69 | #507 |
| 5 | Llama 3.3 70B Instruct | 65.52 | #534 |
| 6 | Qwen 2.5 14B | 65.36 | #725 |
| 7 | tempmotacilla-cinerea-0308 | 65.36 | #395 |
| 8 | AIC-1 | 65.03 | #231 |
| 9 | Qwen 2.5 72B Instruct | 64.54 | #454 |
| 10 | Fanar-1-9B-Instruct | 64.54 | #633 |
| 11 | SILMA-9B-Instruct-v1.0 | 64.54 | #913 |
| 12 | Yehia-7B-preview | 64.05 | #646 |
| 13 | ALLaM-7B-Instruct-preview | 63.73 | #718 |
| 14 | Qwen 2 72B | 63.24 | #521 |
| 15 | Hala-9B | 62.91 | #504 |
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-arabic-language-general · How It Works · Data refreshed daily, snapshot 2026-10-20.