Open Arabic LLM - Alghafa Multiple Choice Rating Sentiment Task: leaderboard
Metric: Accuracy (%). Source: huggingface.co. 163 models tracked.
Top models
| # | Model | Score |
|---|---|---|
| 1 | Llama 3.3 70B Instruct | 58.3 |
| 2 | free-evo-qwen72B-v0.8-re | 57.83 |
| 3 | Qwen 2 72B | 57.66 |
| 4 | aya-23-35B | 57.38 |
| 5 | Qwen 2.5 72B | 55.26 |
| 6 | Qwen 2.5 14B | 55.01 |
| 7 | aya-expanse-32B | 54.98 |
| 8 | Qwen 2.5 32B Instruct | 54.96 |
| 9 | Qwen2.5-32B-Instruct-CFT | 54.96 |
| 10 | Qwen 2.5 7B | 54.95 |
| 11 | Awqward2.5-32B-Instruct | 54.93 |
| 12 | Qwentile2.5-32B-Instruct | 54.93 |
| 13 | oxyge1-33B | 54.65 |
| 14 | Qwen2.5-32B-Instruct-abliterated-v2 | 54.61 |
| 15 | SauerkrautLM-Nemo-12B-Instruct | 54.51 |
Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-alghafa-multiple-choice-rating-sentiment-task · How It Works · Data refreshed daily, snapshot 2026-09-05.