{"id":12629,"date":"2026-06-29T00:00:53","date_gmt":"2026-06-29T07:00:53","guid":{"rendered":"https:\/\/devblogs.microsoft.com\/cosmosdb\/?p=12629"},"modified":"2026-06-30T09:43:42","modified_gmt":"2026-06-30T16:43:42","slug":"spring-ai-2-0-is-ga-vector-search-memory-and-agents-on-azure-cosmos-db","status":"publish","type":"post","link":"https:\/\/devblogs.microsoft.com\/cosmosdb\/spring-ai-2-0-is-ga-vector-search-memory-and-agents-on-azure-cosmos-db\/","title":{"rendered":"Spring AI 2.0 is GA: Vector Search, Memory, and Agents on Azure Cosmos DB"},"content":{"rendered":"

The wait is over.\u00a0Spring AI 2.0 is generally available<\/a><\/strong>, and Azure Cosmos DB is right there with it. With this release, Spring AI graduates into a mature, production-ready framework for building AI applications in Java, and Azure Cosmos DB ships\u00a0dedicated, vendor-maintained integrations<\/strong>\u00a0that plug straight into the Spring AI ecosystem.<\/p>\n

The Spring AI 2.0 GA announcement<\/a> names Azure Cosmos DB among its vendor-maintained modules, maintained directly by Microsoft rather than the core Spring AI team. This means the integration is built and supported by the engineers who work on Cosmos DB itself, bringing deep, first-hand knowledge of how to get the most out of it.<\/p>\n

If you are a Java or Spring developer building RAG pipelines, chatbots, or multi-agent systems, this is a great moment to start (or restart) on Azure Cosmos DB.<\/p>\n

\"Spring<\/a><\/p>\n

What changed in Spring AI 2.0 (and why it’s good news for Cosmos DB)<\/h2>\n

Spring AI 2.0 brings a cleaner, more modular architecture built on\u00a0Spring Boot 4.1<\/strong>\u00a0and\u00a0Spring Framework 7<\/strong>, with a fully\u00a0null-safe (JSpecify)<\/a>\u00a0codebase, Jackson 3 serialization, and stable abstractions for vector stores, chat memory, tool calling, and the\u00a0ChatClient<\/code>\u00a0API. As part of this evolution, vendor-specific integrations now live in their own dedicated repositories rather than the core monorepo.<\/p>\n

For Azure Cosmos DB, that means a\u00a0purpose-built home<\/strong>\u00a0with its own release cadence, changelog, roadmap, and\u00a0documentation site.<\/a> These modules implement Spring AI’s standard interfaces, so you can swap implementations without rewriting application code, while getting the full power of Azure Cosmos DB underneath: global distribution, elastic scale, SLA-backed performance, and a vector index designed for production AI workloads.<\/p>\n

What’s in the box<\/h2>\n

The repository ships four modules, published under the\u00a0com.azure.spring.ai<\/code>\u00a0group ID:<\/p>\n\n\n\n\n\n\n\n\n
Module<\/th>\nWhat it does<\/th>\n<\/tr>\n<\/thead>\n
spring-ai-azure-cosmos-db-store<\/code><\/td>\nVector store backed by Azure Cosmos DB, powered by the\u00a0DiskANN<\/strong>\u00a0index for fast, scalable similarity search<\/td>\n<\/tr>\n
spring-ai-autoconfigure-vector-store-azure-cosmos-db<\/code><\/td>\nZero-config Spring Boot auto-configuration for the vector store<\/td>\n<\/tr>\n
spring-ai-model-chat-memory-repository-cosmos-db<\/code><\/td>\nA\u00a0ChatMemoryRepository<\/code>\u00a0implementation for durable, long-term conversation memory<\/td>\n<\/tr>\n
spring-ai-autoconfigure-model-chat-memory-repository-cosmos-db<\/code><\/td>\nZero-config Spring Boot auto-configuration for chat memory<\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

\ud83d\uddc4\ufe0f Vector search with DiskANN<\/h3>\n

The vector store stores document embeddings in Azure Cosmos DB and serves similarity queries using\u00a0DiskANN<\/strong>, Microsoft Research’s disk-based approximate nearest neighbor algorithm. DiskANN delivers low-latency, high-recall vector search that scales to large datasets, so your retrieval-augmented generation (RAG) workloads stay fast as your data grows, all from the same database that holds your operational data.<\/p>\n

\ud83d\udcac Persistent chat memory<\/h3>\n

The\u00a0CosmosDBChatMemoryRepository<\/code>\u00a0implements Spring AI’s native\u00a0ChatMemoryRepository<\/code>\u00a0interface, giving your agents durable, long-term memory that survives restarts and spans sessions. Conversation history is partitioned by\u00a0\/conversationId<\/code>\u00a0for predictable performance, and the database and container are created automatically, with no manual schema setup required.<\/p>\n

Get started in minutes<\/h2>\n

Add the auto-configuration dependency:<\/p>\n

<dependency<\/span>><\/span>\r\n    <groupId<\/span>><\/span>com.azure.spring.ai<\/groupId<\/span>><\/span>\r\n    <artifactId<\/span>><\/span>spring-ai-autoconfigure-vector-store-azure-cosmos-db<\/artifactId<\/span>><\/span>\r\n    <version<\/span>><\/span>1.0.0<\/version<\/span>><\/span>\r\n<\/dependency<\/span>><\/span>\r\n<\/code><\/pre>\n
Configure your connection in\u00a0application.properties<\/code>:<\/p>\n
spring.ai.vectorstore.cosmosdb.endpoint<\/span>=https:\/\/your-account.documents.azure.com:443\/<\/span>\r\nspring.ai.vectorstore.cosmosdb.databaseName<\/span>=my-database<\/span>\r\nspring.ai.vectorstore.cosmosdb.containerName<\/span>=my-vectors<\/span>\r\nspring.ai.vectorstore.cosmosdb.vectorDimensions<\/span>=1536<\/span>\r\n<\/code><\/pre>\n
Then inject and use the standard Spring AI\u00a0VectorStore<\/code>:<\/p>\n
@Autowired<\/span>\r\nprivate<\/span> VectorStore vectorStore;\r\n\r\n\/\/ Add documents<\/span>\r\nvectorStore.add(List.of(new<\/span> Document<\/span>(\"Hello world\"<\/span>)));\r\n\r\n\/\/ Search<\/span>\r\nList<Document> results = vectorStore.similaritySearch(\r\n    SearchRequest.builder().query(\"Hello\"<\/span>).topK(5<\/span>).build());\r\n<\/code><\/pre>\n
Wiring up chat memory is just as simple:<\/p>\n
@Autowired<\/span>\r\nCosmosDBChatMemoryRepository chatMemoryRepository;\r\n\r\nChatMemory<\/span> chatMemory<\/span> =<\/span> MessageWindowChatMemory.builder()\r\n    .chatMemoryRepository(chatMemoryRepository)\r\n    .maxMessages(10<\/span>)\r\n    .build();\r\n<\/code><\/pre>\n
That’s it. Spring Boot auto-configuration handles client creation, container provisioning, and credential resolution.<\/p>\n
Built for production<\/h2>\n
This release was tuned for real-world Spring Boot 4.x deployments:<\/p>\n