Open Arabic LLM - Alghafa Multiple Choice Grounded Statement Soqal Task Leaderboard — The Aggregate

Open Arabic LLM - Alghafa Multiple Choice Grounded Statement Soqal Task — leaderboard

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

Top models

#	Model	Score
1	lambda-qwen2.5-32B-dpo-test	95.33
2	Qwen 2.5 32B Instruct	94.67
3	Awqward2.5-32B-Instruct	94.67
4	Qwentile2.5-32B-Instruct	94.67
5	Qwen2.5-32B-Instruct-CFT	94.67
6	Rombos-LLM-V2.6-Qwen-14B	94.67
7	lambda-qwen2.5-14B-dpo-test	94
8	recoilme-gemma-2-9B-v0.4	94
9	Qwen2.5-14B-Gutenberg-1e-Delta	94
10	Qwen 2.5 72B Instruct	93.33
11	Qwen 2.5 14B Instruct	93.33
12	Saka-14B	93.33
13	Qwen2.5-Lumen-14B	93.33
14	calme-2.1-qwen2.5-72B	93.33
15	calme-2.2-qwen2.5-72B	93.33

Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-alghafa-multiple-choice-grounded-statement-soqal-task · How the rankings work · Data refreshed daily, snapshot 2026-07-22.