Open Arabic LLM - Arabic MMLU Arabic Language (High School): leaderboard
The "Arabic MMLU Arabic Language (High 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: Around 2031. 163 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | Saka-14B | 46.92 | #344 |
| 2 | Qwen2.5-32B-Instruct-abliterated-v2 | 46.15 | #394 |
| 3 | Fanar-1-9B-Instruct | 45.9 | #633 |
| 4 | free-evo-qwen72B-v0.8-re | 45.38 | #435 |
| 5 | gemma-3-27B-pt | 44.36 | #502 |
| 6 | Qwen 2.5 72B Instruct | 44.1 | #454 |
| 7 | ALLaM-7B-Instruct-preview | 44.1 | #718 |
| 8 | Yehia-7B-preview | 43.59 | #646 |
| 9 | Qwen 2.5 14B | 43.33 | #725 |
| 10 | AceGPT-v2-32B-Chat | 43.08 | #507 |
| 11 | tempmotacilla-cinerea-0308 | 43.08 | #395 |
| 12 | gemma-3-12B-pt | 42.82 | #686 |
| 13 | AIC-1 | 42.56 | #231 |
| 14 | jais-adapted-70B-chat | 42.31 | #636 |
| 15 | Llama 3.3 70B Instruct | 41.54 | #534 |
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-high-school · How It Works · Data refreshed daily, snapshot 2026-10-20.