ResearchBenchmarksMultitask Language Understanding on MMLU-pl Global (test)Follow43.47AccuracyQwen325.561230.210634.8639.5094Dec 25, 2025Evaluation ResultsMethodMethodLinksAccuracyQwen32025.1243.47Qwen2.52025.1242Gamayun2025.1240.47Llama3.22025.1234.8Gemma32025.1234.75EuroLM2025.1226.25