Open Arabic LLM - Aratrust Privacy — leaderboard
Metric: Accuracy (%). Source: huggingface.co. 162 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instruct | 98.25 |
| 2 | lambda-qwen2.5-14B-dpo-test | 98.25 |
| 3 | Qwen 2 72B | 98.25 |
| 4 | recoilme-gemma-2-9B-v0.4 | 98.25 |
| 5 | Qwen 2.5 14B Instruct | 96.49 |
| 6 | Qwen 2.5 72B | 96.49 |
| 7 | Qwen2.5-32B-Instruct-abliterated-v2 | 96.49 |
| 8 | Qwentile2.5-32B-Instruct | 96.49 |
| 9 | EZO-Qwen2.5-32B-Instruct | 96.49 |
| 10 | free-evo-qwen72B-v0.8-re | 96.49 |
| 11 | Qwen2.5-14B-Gutenberg-1e-Delta | 96.49 |
| 12 | Qwen2.5-Lumen-14B | 96.49 |
| 13 | Rombos-LLM-V2.6-Qwen-14B | 96.49 |
| 14 | Qwen 2.5 72B Instruct | 94.74 |
| 15 | Phi-4 Mini Instruct | 94.74 |
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-aratrust-privacy · How the rankings work · Data refreshed daily, snapshot 2026-07-22.