Open Arabic LLM - Arabic MMLU HT High School Mathematics — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 163 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Ultiima-72B | 55.56 |
| 2 | Rombos-LLM-V2.5-Qwen-72B | 55.19 |
| 3 | Saka-14B | 49.63 |
| 4 | gemma-3-27B-pt | 47.78 |
| 5 | Qwen3-8B-Base | 46.67 |
| 6 | Qwen 2.5 14B | 45.19 |
| 7 | calme-2.2-qwen2.5-72B | 44.44 |
| 8 | Qwen 2 72B | 44.07 |
| 9 | calme-2.1-qwen2.5-72B | 44.07 |
| 10 | lambda-qwen2.5-32B-dpo-test | 43.33 |
| 11 | Qwen2.5-32B-Instruct-abliterated-v2 | 42.96 |
| 12 | Qwen 2.5 7B | 41.48 |
| 13 | tempmotacilla-cinerea-0308 | 41.11 |
| 14 | Qwen 2.5 72B Instruct | 40.74 |
| 15 | Arcee-Spark | 40.74 |
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-arabic-mmlu-ht-high-school-mathematics · How the rankings work · Data refreshed daily, snapshot 2026-07-22.