Alibaba Cloud Model Studio RAG Knowledge Base documentation.
Connect your private data to large language models with Retrieval-Augmented Generation (RAG). Upload your documents, and the platform handles parsing, chunking, embedding, and indexing automatically — then exposes retrieval and Streaming Q&A through the API.
Quickstart
Build your first Knowledge Base in five steps, then run a retrieval and a Q&A call.
Core Concepts
How RAG, chunking, embeddings, indexing, and retrieval fit together.
Create a Knowledge Base
Pick a Knowledge Base type and configure parsing and Chunking Strategy.
Data Connectors
Ingest from files, tables, MySQL, OSS, and more.
Knowledge Retrieval
Multi-index Retrieval with agentic multi-turn search.
Knowledge Q&A
Streaming RAG Q&A backed by a large language model.
Integration Channels
API, MCP, CLI, and Skill integrations.
API Reference
28 interactive API references with an in-browser test console.