Open Arabic LLM - Arabic MMLU Driving Test — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 163 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Ultiima-72B | 79.93 |
| 2 | Rombos-LLM-V2.5-Qwen-72B | 79.69 |
| 3 | Qwen 2.5 72B Instruct | 77.21 |
| 4 | calme-2.2-qwen2.5-72B | 76.88 |
| 5 | Qwen 2 72B | 76.8 |
| 6 | calme-2.1-qwen2.5-72B | 76.71 |
| 7 | gemma-3-27B-pt | 76.14 |
| 8 | Saka-14B | 75.89 |
| 9 | Gemma 3 27B (IT) | 75.47 |
| 10 | Qwentile2.5-32B-Instruct | 74.81 |
| 11 | tempmotacilla-cinerea-0308 | 73.82 |
| 12 | Qwen3-8B-Base | 73.33 |
| 13 | Qwen2.5-32B-Instruct-abliterated-v2 | 73.33 |
| 14 | free-evo-qwen72B-v0.8-re | 73.25 |
| 15 | lambda-qwen2.5-32B-dpo-test | 73.25 |
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-arabic-mmlu-driving-test · How the rankings work · Data refreshed daily, snapshot 2026-07-22.