Open Arabic LLM - Arabic MMLU HT High School Biology — leaderboard

Metric: Accuracy (%). Source: huggingface.co. 163 models tracked.

Top models

#ModelScore
1Ultiima-72B81.96
2Rombos-LLM-V2.5-Qwen-72B81.9
3calme-2.1-qwen2.5-72B81.72
4calme-2.2-qwen2.5-72B81.62
5Qwen 2.5 72B Instruct80.75
6Qwen 2 72B77.1
7Llama 3.3 70B Instruct75.37
8Qwen 2.5 72B74.98
9gemma-3-27B-pt74.35
10Gemma 3 27B (IT)73.64
11lambda-qwen2.5-32B-dpo-test73.01
12Awqward2.5-32B-Instruct72.65
13Qwen 2.5 32B Instruct72.62
14Qwen2.5-32B-Instruct-CFT72.49
15Qwentile2.5-32B-Instruct72.36

Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-arabic-mmlu-ht-high-school-biology · How the rankings work · Data refreshed daily, snapshot 2026-07-22.