Spring AI Spring Boot Auto Configuration

Spring AI Spring Boot Auto Configuration provides automatic setup and configuration for Spring AI components, including AI model providers, vector stores, Model Context Protocol (MCP) support, and infrastructure components. This module enables zero-configuration Spring Boot applications to use AI capabilities by automatically creating and configuring beans based on classpath detection and application properties.

Package Information

• Package Group: org.springframework.ai
• Artifact ID: spring-ai-spring-boot-autoconfigure
• Package Type: Maven (Multi-module)
• Language: Java
• Version: 1.1.2
• Module Count: 56 autoconfiguration modules
• Installation: Add specific autoconfiguration modules as Maven dependencies

Quick Start

1. Add Dependencies:

<dependency>
    <groupId>org.springframework.ai</groupId>
    <artifactId>spring-ai-openai-spring-boot-starter</artifactId>
    <version>1.1.2</version>
</dependency>

2. Configure Properties:

spring.ai.openai.api-key=your-api-key
spring.ai.openai.chat.options.model=gpt-4
spring.ai.openai.chat.options.temperature=0.7

3. Use Autoconfigured Beans:

@Service
public class ChatService {
    private final ChatModel chatModel;
    
    public ChatService(ChatModel chatModel) {
        this.chatModel = chatModel;
    }
    
    public String chat(String message) {
        return chatModel.call(message);
    }
}

See Full Quick Start Guide →

Architecture Overview

Spring AI autoconfiguration is organized into five major categories:

Module Categories

Total: 56 autoconfiguration modules

Autoconfiguration Mechanism

Each module uses Spring Boot's @AutoConfiguration mechanism:

• Conditional Activation: Modules activate based on classpath presence
• Property-based Configuration: Enable/disable via spring.ai.* properties
• Bean Creation: Automatic creation when conditions are met
• Customization Support: Override via @ConditionalOnMissingBean
• Ordering Control: Uses @AutoConfiguration(before/after) for proper initialization

Core Capabilities

1. Common Infrastructure

Retry capabilities with exponential backoff and intelligent error classification.

spring.ai.retry.max-attempts=10
spring.ai.retry.backoff.initial-interval=2s
spring.ai.retry.backoff.multiplier=5

Common Module Documentation →

2. Model Context Protocol (MCP)

Client and server support with multiple transports (stdio, SSE, HTTP).

spring.ai.mcp.client.enabled=true
spring.ai.mcp.client.stdio.connections.myserver.command=node
spring.ai.mcp.client.stdio.connections.myserver.args[0]=server.js

MCP Client Documentation → | MCP Server Documentation →

3. Chat Infrastructure

Chat clients, memory management, observation, and tool calling.

spring.ai.chat.client.enabled=true
spring.ai.chat.memory.repository.mongo.create-indices=true
spring.ai.chat.observations.log-prompt=true

Chat Infrastructure Documentation →

4. AI Model Providers

Support for 18 AI service providers with standard interfaces.

Supported Providers: OpenAI, Anthropic, Azure OpenAI, AWS Bedrock, Google Gemini, Ollama, Mistral AI, Vertex AI, and 11 more.

# OpenAI
spring.ai.openai.api-key=your-key
spring.ai.openai.chat.options.model=gpt-4

# Anthropic
spring.ai.anthropic.api-key=your-key
spring.ai.anthropic.chat.options.model=claude-3-5-sonnet-20241022

AI Providers Documentation →

5. Vector Stores

Support for 20 vector databases with common interface.

Supported Stores: Pinecone, Weaviate, Milvus, Qdrant, PGVector, Chroma, MongoDB Atlas, Redis, and 12 more.

# PGVector
spring.ai.vectorstore.pgvector.initialize-schema=true
spring.ai.vectorstore.pgvector.index-type=HNSW

# Pinecone
spring.ai.vectorstore.pinecone.api-key=your-key
spring.ai.vectorstore.pinecone.index-name=my-index

Vector Stores Documentation →

Common Configuration Patterns

Enabling/Disabling Modules

spring.ai.mcp.client.enabled=false
spring.ai.retry.enabled=false
spring.ai.chat.client.enabled=false

Retry Configuration

spring.ai.retry.max-attempts=5
spring.ai.retry.backoff.initial-interval=1s
spring.ai.retry.backoff.multiplier=2
spring.ai.retry.on-http-codes=429,503
spring.ai.retry.exclude-on-http-codes=401,403

Observation and Monitoring

spring.ai.chat.observations.log-prompt=true
spring.ai.chat.observations.log-completion=true
management.metrics.export.prometheus.enabled=true
management.tracing.enabled=true

Spring Cloud Bindings

Automatic service binding integration for cloud environments (Kubernetes, Cloud Foundry).

Supported Bindings: MCP Client, Ollama, Chroma, GemFire, Milvus, OpenSearch, Qdrant, Typesense, Weaviate.

See Spring Cloud Bindings Details →

AOT and GraalVM Native Image

Built-in support for native images with automatic runtime hints.

./mvnw -Pnative native:compile
./target/my-spring-ai-app

See Native Image Details →

Module Organization

Complete Module List

Common (1 module)

• spring-ai-autoconfigure-retry

MCP (6 modules)

• spring-ai-autoconfigure-mcp-client-common
• spring-ai-autoconfigure-mcp-client-httpclient
• spring-ai-autoconfigure-mcp-client-webflux
• spring-ai-autoconfigure-mcp-server-common
• spring-ai-autoconfigure-mcp-server-webflux
• spring-ai-autoconfigure-mcp-server-webmvc

Chat Infrastructure (11 modules)

• spring-ai-autoconfigure-model-chat-client
• spring-ai-autoconfigure-model-chat-memory
• spring-ai-autoconfigure-model-chat-memory-repository-* (5 variants)
• spring-ai-autoconfigure-model-chat-observation
• spring-ai-autoconfigure-model-embedding-observation
• spring-ai-autoconfigure-model-image-observation
• spring-ai-autoconfigure-model-tool

AI Providers (18 modules)

• Anthropic, Azure OpenAI, AWS Bedrock, DeepSeek, ElevenLabs, Google Gemini, Hugging Face, MiniMax, Mistral AI, OCI GenAI, Ollama, OpenAI, OpenAI SDK, PostgresML, Stability AI, Transformers, Vertex AI, Zhipu AI

Vector Stores (20 modules)

• Azure AI Search, Azure Cosmos DB, Cassandra, Chroma, Couchbase, Elasticsearch, GemFire, MariaDB, Milvus, MongoDB Atlas, Neo4j, OpenSearch, Oracle, PGVector, Pinecone, Qdrant, Redis, Typesense, Weaviate, Vector Store Observation

See Complete Module Details →

Configuration Reference

Quick Configuration Table

See Complete Configuration Reference →

Resources

Guides

• Quick Start Guide - Get started in 5 minutes
• Retry Configuration Guide - Configure retry behavior
• MCP Integration Guide - Integrate with MCP servers
• RAG Pattern Guide - Build retrieval-augmented generation

Examples

• Real-World Scenarios - Production use cases
• Edge Cases - Advanced scenarios and troubleshooting
• Multi-Provider Setup - Using multiple AI providers
• Complete Applications - Full application examples

Reference

• Common API Reference - Retry and error handling APIs
• MCP Client API Reference - MCP client APIs
• MCP Server API Reference - MCP server APIs
• Chat Infrastructure API Reference - Chat APIs
• Provider API Reference - AI provider APIs
• Vector Store API Reference - Vector store APIs
• Configuration Properties - All properties
• Spring Cloud Bindings - Service binding
• Native Image Support - AOT/GraalVM
• Custom Conditions - Advanced conditions
• Type Converters - Property converters
• Troubleshooting - Common issues

Key Features

✅ Zero Configuration - Automatic bean creation based on classpath and properties
✅ Provider Agnostic - Consistent interfaces across 18 AI providers
✅ Production Ready - Retry, observation, and error handling built-in
✅ Flexible - Easy customization via properties or custom beans
✅ Cloud Native - Spring Cloud Bindings and native image support
✅ Comprehensive - 20 vector stores, 11 chat infrastructure modules, 6 MCP modules

Additional Resources

• Spring AI Documentation: https://docs.spring.io/spring-ai/reference/
• Spring Boot Autoconfiguration: https://docs.spring.io/spring-boot/docs/current/reference/html/using.html#using.auto-configuration
• Model Context Protocol: https://modelcontextprotocol.io/

Summary

Spring AI Spring Boot Auto Configuration provides comprehensive, zero-configuration setup for AI capabilities in Spring Boot applications. With 56 autoconfiguration modules covering chat models, embeddings, vector stores, MCP, and infrastructure components, it enables rapid development of AI-powered applications while maintaining flexibility for customization and production deployment.