langchain-rust: Build LLM apps with Ollama + local models in pure Rust — no Python needed

AI 中文解读

langchain-rust来了！这套纯Rust编写的AI框架，让你彻底摆脱Python环境，直接用Ollama跑本地大模型，一条命令就能搞定。通俗点说，以前想用AI开发应用，得先装Python、配虚拟环境、处理各种依赖冲突，麻烦得很。现在用Rust，编译成单个文件，不需要“虚拟环境”这些繁琐步骤，加了就能跑。它内置的“知识库搜索”能同时用关键词和向量两种方式找资料，英文中文都能处理，还支持把知识图谱和联网搜索结合，让AI回答得更准、更不容易“一本正经地胡说八道”。甚至代码执行、记忆管理、安全护栏这些高级功能都打包好了。对普通人来说，这意味着将来会有更多轻快、流畅的AI小工具出现——不用联网也能跑，隐私更安全，手机或电脑上直接执行。对开发者而言，写AI应用像写普通软件一样简单，生态门槛大大降低，未来可能涌现出一大批好用又便宜的本地方案，你的日常办公、学习助手也会变得更聪明、更贴心。

If you're running local models through Ollama and tired of Python's overhead, check out <a href="https://github.com/atliliw/langchainrust" rel="noopener noreferrer">langchain-rust</a>. It's a full LLM framework in pure Rust that works great with local models: <ul> <li> Ollama support — first-class integration with tool calling, vision, and streaming</li> <li> 9 vector store backends — InMemory, SQLite, Qdrant, ChromaDB, Redis, PGVector, MongoDB, Pinecone, FileVectorStore</li> <li> BM25 keyword search — with Chinese/English tokenization, no external dependency</li> <li> Hybrid retrieval — BM25 + Vector with RRF fusion for better recall</li> <li> GraphRAG — Knowledge graph construction + community detection, all local</li> <li> CorrectiveRAG — Self-correcting retrieval with hallucination detection</li> <li> Code Interpreter — LocalSandbox (subprocess), E2B cloud, or WASM sandbox</li> <li> LocalEmbeddings — Run embeddings without calling an API</li> </ul> Plus: LangGraph workflows, MCP client/server, 7 memory types, guardrails, and 12+ built-in tools. Single binary, no virtualenv, no pip conflicts. Just <code>cargo add langchainrust</code> and go. GitHub: <a href="https://github.com/atliliw/langchainrust" rel="noopener noreferrer">https://github.com/atliliw/langchainrust</a> Docs: <a href="https://docs.rs/langchainrust" rel="noopener noreferrer">https://docs.rs/langchainrust</a>