Global

This repository is a comprehensive collection of 132 projects, tutorials, and recipes for building powerful LLM-powered applications, including text agents, voice assistants, RAG apps, and MCP-backed tools.

🔍大模型应用开发实战一：RAG 技术全栈指南，在线阅读地址：

Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.

🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.

A community-driven hub of 42+ runnable notebooks covering RAG techniques from foundational to cutting-edge - the intuition, the code, and the references to build more accurate, context-rich retrieval systems.

[MLsys2026 Best Paper]: RAG on Everything with LEANN. Enjoy 97% storage savings while running a fast, accurate, and 100% private RAG application on your personal device.