galileo-ai/agent-leaderboard

Agent Leaderboard是一个综合性的评价语言模型在复杂场景中有效利用工具的能力的基准数据集。它包含了多个领域和交互类型的数据，用于评估语言模型在选择和使用工具以满足实际世界需求方面的表现。

Agent Leaderboard is a benchmark dataset designed to evaluate the ability of language models to effectively utilize tools in complex scenarios. It encompasses data from multiple domains and interaction types to assess how well language models select and use tools based on real-world requirements.