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

The "Alghafa Multiple Choice Grounded Statement Soqal Task" subtask of Open Arabic LLM. A task from the Open Arabic LLM Leaderboard, evaluating models on Arabic-language knowledge, reasoning and understanding across native and translated benchmarks.

Metric: Accuracy (%). Source: huggingface.co. Saturation forecast: Estimated already saturated. 163 models tracked.

Top models

#ModelScoreOverall rank
1AIC-196.67#237
2Qwen 2.5 32B Instruct94.67#511
3Qwentile2.5-32B-Instruct94.67#345
4Awqward2.5-32B-Instruct94.67#370
5Qwen2.5-32B-Instruct-CFT94.67#377
6Rombos-LLM-V2.6-Qwen-14B94.67#522
7lambda-qwen2.5-14B-dpo-test94#482
8Hala-9B94#518
9Qwen2.5-14B-Gutenberg-1e-Delta94#473
10recoilme-gemma-2-9B-v0.494#575
11Qwen 2.5 72B Instruct93.33#470
12Qwen 2.5 14B Instruct93.33#655
13AceGPT-v2-32B-Chat93.33#519
14Saka-14B93.33#352
15Qwen2.5-Lumen-14B93.33#483

No result here: #3 Claude Opus 5.5, #5 GPT-6 Astra, #6 GPT-6.1 Sol.

Interactive version: theaggregate.ai/benchmark?slug=open-arabic-llm-alghafa-multiple-choice-grounded-statement-soqal-task · How It Works · Data refreshed daily, snapshot 2026-11-06.