ResearchBenchmarksMulti-task language understanding on Open Arabic LLM LeaderboardFollow49.8AccuracyGamayun37.73640.8684447.132Dec 25, 2025Evaluation ResultsMethodMethodLinksAccuracyGamayunscenario=zero-shotscenario=zero-shot2025.1249.8Gemma3scenario=zero-shotscenario=zero-shot2025.1243Qwen2.5scenario=zero-shotscenario=zero-shot2025.1242.7Qwen3scenario=zero-shot, th...scenario=zero-shot, thinking_mode=no-thinking2025.1241.5EuroLMscenario=zero-shotscenario=zero-shot2025.1238.9Llama3.2scenario=zero-shotscenario=zero-shot2025.1238.2