ResearchBenchmarksComplex Reasoning and Instruction Following on ItaBench leaderboard tasksFollow37.88BBH-it ScoreQwen2.529.66431.79733.9336.063Dec 25, 2025Evaluation ResultsMethodMethodLinksBBH-it ScoreGPQA-it ScoreMUSR-it ScoreIFEval-it (avg)MMLU-pro-it ScoreQwen2.52025.1237.8826.9137.9628.6923.39Gamayun2025.1236.7430.1136.7737.4821.03Qwen32025.1236.6127.5940.3424.2226.13Llama3.22025.1233.2426.5836.7728.9817.48Gemma32025.1231.6126.6635.4541.7313.2EuroLM2025.1229.9824.7337.9626.1810.26