📄 中文摘要
单一模型的 AI 系统在生产环境中逐渐显得不足。当前最强大的 AI 系统通常结合了多个模型、检索器、验证器和工具，以协同工作。复合 AI 系统通过将任务的不同部分路由到最合适的组件，而不是将所有内容发送到单一模型，从而提高效率。这些组件通常包括多个语言模型（针对不同子任务的不同模型）、检索系统（向量数据库、搜索等）等。通过这种架构，可以更有效地处理复杂的任务，提升 AI 产品的整体性能。
复合 AI 系统：如何将多个模型连接到单一生产产品中
📄 English Summary
Compound AI Systems: How I Connect Multiple Models in a Single Production Product
Single-model AI systems are increasingly inadequate for production environments. The most capable AI systems today integrate multiple models, retrievers, validators, and tools to work collaboratively. A compound AI system routes different parts of a task to the most appropriate component instead of sending everything to a single model, enhancing efficiency. These components typically include multiple language models (different models for different subtasks) and retrieval systems (vector databases, search, etc.). This architecture allows for more effective handling of complex tasks, improving the overall performance of AI products.