Pocket Flow 是一个仅有100行代码的极简LLM框架，旨在简化大型语言模型应用的开发。它利用核心图抽象，使用户能够轻松实现多智能体、工作流和RAG等流行设计模式，且无冗余或依赖。
PocketFlow 当前归类于 RAG / Knowledge Base，更适合那些希望围绕具体工作流而不是单一功能点来选型的团队。从当前资料来看，它尤其强调 轻量级和极简主义（100行代码，零依赖）。、构建和编排AI智能体（例如，研究智能体、多智能体）。 这类能力。如果你关心许可证与可二次使用性，当前页面记录的许可证是 MIT。同时它也具备一定社区热度，当前 GitHub Stars 为 11.0k。
100-line LLM framework is the right size for understanding what's actually happening under the hood.
Production use is possible but the real value is understanding agent patterns without framework magic.
Used as a teaching tool for LLM agent concepts. The small surface area makes it genuinely learnable.
Letting Agents build Agents is a compelling meta-capability when the framework is this minimal.
The simplicity is the feature. Most frameworks hide too much — PocketFlow makes tradeoffs explicit.
