Building a RAG System: Practice Based on Qwen3 Embedding

RAGFlow is an open source RAG (Retrieval-Augmented Generation) engine based on deep document understanding. When integrated with LLM, it provides true question-and-answer capabilities, backed by reliable references from data in a variety of complex formats. 

---

本教程采用算力资源为「双卡 RTX 6000」，具体包含以下内容：

一、启动 RAGFlow 服务器（见 README.md 文件）
二、创建一个知识库（见 Knowledge.md 文件）
三、根据您的知识库建立 AI 聊天（见 Chat.md 文件）
四、开启第一个 Agent（见 Agent.md 文件）
五、交流探讨

---

1. Start the RAGFlow server

This tutorial has configured the RAGFlow service. You do not need to deploy it yourself. After starting the container, you can directly click the API address to access it.

注意：模型文件较大，稍等容器启动 5-6 分钟后访问，否则将显示「Bad Gataway」！

The login information is as follows:

• 👤 username:admin@123.com
• 🔑 password:123456

 —

2. Create a knowledge base

In this step, you can upload the file to the knowledge base in RAGFlow and parse it into a data set. To facilitate the testing of this tutorial, the corresponding knowledge base has been pre-configured for the Agent and Chat tutorials. Users can directly use the default configuration or upload and customize the knowledge base according to the following process.

📖 learn more
ClickHereView the full how-to guide.

---

3. Build AI chat based on knowledge base

The conversation in RAGFlow is based on one or more specific knowledge bases. Once the knowledge base is created and the file parsing is complete, the AI conversation can begin.

📖 learn more
Click 👉 HereView the full how-to guide.

---

4. Start the first Agent

Proxy and RAG are complementary technologies that enhance each other in business applications. **RAGFlow v0.8.0 introduces a proxy mechanism, **which can better meet user needs by configuring different proxies.

📖 learn more
Click 👉 HereView the full how-to guide.

5. Discussion

🖌️ If you see a high-quality project, please leave a message in the background to recommend it! In addition, we have also established a tutorial exchange group. Welcome friends to scan the QR code and remark [SD Tutorial] to join the group to discuss various technical issues and share application effects↓