Spring AI Alibaba's cookbook: runnable samples for every AI feature you'd want

What it does This repo is the examples companion to the Spring AI Alibaba framework. Each subdirectory is a self-contained Spring Boot module demonstrating one capability: basic chat with advisors, image generation, RAG with a vector database, tool calling, and MCP integration. The top-level README acts as an index, with a quickstart matrix mapping each module to its purpose, required services, and environment variables.

The interesting bit It’s not just DashScope demos — the examples cover a spread of providers (OpenAI-compatible endpoints, vLLM, DeepSeek, MiniMax, ZhipuAI), which makes it a decent reference for how Spring AI’s provider abstraction looks in practice across the Chinese and Western model ecosystems.

Key highlights

• Covers the standard AI-app checklist: chat, image generation, RAG, tool calling, MCP, plus memory (mem0) and evaluation examples
• Quickstart matrix documents required services and env vars per module
• Bilingual (Chinese/English) READMEs
• Community contributions welcome

Caveats

• It’s glue code by design — the value is in the patterns, not the engineering
• Some modules lack a .env.example; the README tells you to dig into each module’s application.yml instead
• RAG, MCP, and vector-db examples expect local middleware or containers running first

Verdict Worth a skim if you’re a Java developer evaluating Spring AI Alibaba or porting an AI feature into an existing Spring stack. If you’re not in the Spring ecosystem, there’s little here for you.