Open Arabic LLM - Aratrust Ethics — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 162 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Qwen2.5-32B-Instruct-abliterated-v2 | 81.67 |
| 2 | Awqward2.5-32B-Instruct | 81.67 |
| 3 | tempmotacilla-cinerea-0308 | 81.67 |
| 4 | Qwen2.5-32B-Instruct-CFT | 81.67 |
| 5 | Qwen 2.5 72B Instruct | 80 |
| 6 | gemma-3-12B-pt | 80 |
| 7 | Qwen 2.5 32B Instruct | 80 |
| 8 | Qwen 2.5 72B | 80 |
| 9 | Qwentile2.5-32B-Instruct | 80 |
| 10 | EZO-Qwen2.5-32B-Instruct | 80 |
| 11 | Qwen 2 72B | 80 |
| 12 | Rombos-LLM-V2.5-Qwen-72B | 80 |
| 13 | Ultiima-72B | 80 |
| 14 | Llama 3.3 70B Instruct | 78.33 |
| 15 | aya-expanse-32B | 78.33 |
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-aratrust-ethics · How the rankings work · Data refreshed daily, snapshot 2026-07-22.