⚠️ Upgrading Notes

• If you are using the Anthropic Skills API directly, review your code for the renamed helper classes and update accordingly #5299
• If you are using the Couchbase vector store, update your import statements to reference the new dedicated package location. #5233

⭐ New Features

• Ollama embedding models now support configuring the dimensions parameter, allowing control over the size of generated embeddings 4bd83ba
• JSON schema can now be customized, providing more flexibility for structured output generation 3119086
• Default API endpoint paths are now configured for OpenAI's content moderation API, simplifying integration setup #5300
• Default API endpoint paths are now configured for OpenAI's audio API (speech and transcription), making it easier to use audio capabilities f7eeaf9
• Added support for using Amazon S3 as a vector store backend for storing and retrieving embeddings cd86d38
• Integrated Infinispan as a new vector store option, providing distributed in-memory data grid capabilities for embedding storage 9b105b8
• Added support for Amazon Bedrock Knowledge Base as a vector store, enabling integration with AWS's managed knowledge base service 21c3834
• Enhanced the Redis semantic cache module with improved functionality and performance 7233e8d
• Made the content field name configurable in QdrantVectorStore, allowing customization of the field used to store document content f65756e
• Added a convenience method to the Prompt class for easily retrieving system messages 8a79390
• Added native support for structured outputs using JSON schema validation in the Mistral AI chat model, enabling type-safe responses with automatic schema enforcement. 8757c5b
• OpenAI moderation, audio speech, and audio transcription API paths are now configurable, allowing customization for proxy setups or alternative OpenAI-compatible endpoints. 1d38aec
• Migrated to use Spring Boot 4's native MongoDB Atlas container support for improved testing and development experience. 5b39fbb
• Introduced Mcp*ServerCustomizer interfaces to allow customization of MCP server configurations #5261
• Added new Redis-based semantic cache advisor implementation, enabling intelligent caching of semantically similar queries to improve performance and reduce API calls 391cbbb
• OllamaChatOptions now implements StructuredOutputChatOptions with corresponding methods, unit tests, and integration tests. Enhanced JSON format handling for Ollama chat options 163df2e
• Improved model provider detection logic for Microsoft Foundry integration, providing better model identification and configuration support 5c18a40
• Add support for dynamic tool schema enhancement, allowing runtime modification of tool argument definitions for more flexible function calling 3c4b6dc
• Add conversationHistoryEnabled option to ToolCallAdvisor, providing finer control over whether conversation history is included in tool call contexts 5ad8485
• Add JSpecify annotations to commons module for improved null-safety analysis and compile-time verification #5101

🪲 Bug Fixes

• Resolved an issue where ToolCallAdvisor would not preserve system messages when memory functionality was disabled 1dac9d0
• Fixed incorrect observation provider usage in the Google GenAI embedding model implementation #5227
• Corrected conditional logic in Infinispan AutoConfiguration to ensure proper bean creation #5295
• Resolved issues in the MistralAiChatModelIT.chatClientEntityWithStructuredOutput integration test to ensure proper validation of structured output functionality. cf2212b
• Corrected code style violations in the spring-ai-redis-semantic-cache module to maintain code quality standards. 0108011
• Fixed MCP auto-configuration to work properly in non-web application contexts #5261
• Improved exception handling in the ChromaApi for better error reporting e291712
• Fixed failing Ollama integration tests by modifying tool descriptions to better guide the model in providing geographical coordinates 864487f
• Enhanced JdbcChatMemoryRepositorySchemaInitializer for improved database schema setup d202db1
• Corrected string parsing issues in the grammar implementation 8ecc334
• Corrected invalid Filter.Expression constructor usage in documentation examples #5243
• Resolved checkstyle violations in Redis auto-configuration code to maintain code quality standards 07ccd2c
• Resolved token count inconsistencies in Ollama integration tests that occurred with varying model versions c2dfedb
• Updated Anthropic integration tests to use the currently active model for improved test reliability 5982176
• Added explicit null safety check to prevent potential NullPointerException in Neo4jChatMemory MediaBuilder e4c2fb4
• Fixed missing dependency artifacts for Redis chat memory repository in the Bill of Materials, ensuring proper dependency management for projects using Redis-based chat memory. 81427bb
• Fixed an issue where the Neo4jVectorStore's doAdd method was ignoring the sessionConfig parameter, ensuring proper session configuration handling during document addition. #5200 via #5191
• Fixed an issue where chat memory advisors could not properly handle ToolResponseMessage types, improving compatibility with tool-based conversations #5174
• Resolve build issues with Antora documentation generation 95bf79a
• Corrected exception message validation in ChromaDB HTTP client/server communication 5dcac7b

📓 Documentation

• Updated documentation for Prompt-related functionality 190fe98
• Added documentation for the configurable content_field_name feature in Qdrant vector store 002a208
• Added documentation for the new configuration properties that allow customization of OpenAI speech, transcription, and moderation API endpoints. f8ad209
• Updated documentation to cover the native structured output converter functionality for Mistral AI models. a9f49a0
• Added comprehensive documentation for the new Redis semantic cache feature, including configuration and usage examples 77e115f
• Corrected duplicated step numbering in RetrievalAugmentationAdvisor documentation for improved clarity #5160
• General typo corrections throughout documentation 64de839
• Corrected typographical errors in the Gemini integration documentation. 5e397b5
• Added comprehensive documentation guide for using LLM as a Judge pattern for evaluating AI model outputs #5145
• Added documentation guide explaining how to implement dynamic tool discovery for runtime function calling capabilities #5141
• Add reference documentation for Anthropic Custom Skills integration ba49f4e
• Remove redundant /v1 suffix from OpenAI base-url examples in documentation for clearer configuration guidance d281da4
• Update incorrect comment in chat client code for accuracy #4891
• Corrected code syntax in chat options builder documentation example #4915
• Corrected typo in error message for ChromaDB collection existence validation #5080
• Updated OpenAI embedding model documentation reference in TokenCountBatchingStrategy Javadoc to point to current documentation #5083

🔨 Dependency Upgrades

• Updated the Model Context Protocol (MCP) SDK dependency to version 0.17.2 d81e47d
• Updated Spring Boot dependency to version 4.0.1 5f1e021
• Updated Mistral AI chat model enumeration to reflect latest available models b77202c
• Updated OpenAI Java SDK dependency to version 4.13.0, bringing latest improvements and compatibility updates #5187

🔩 Build Updates

• Updated static analysis tools NullAway and Error Prone to their latest versions for improved code quality checks d232c82
• Migrated S3 Vector Store to use JSpecify nullability annotations for improved null safety 24aeed5
• Made ResponseErrorHandler configurable as an optional bean for more flexible error handling daa08c8
• Added JSpecify nullability annotations to document-readers modules to improve null-safety and IDE support. #5287
• Added JSpecify nullability annotations to the RedisSemanticCache module for better null-safety guarantees. #5284
• Added JSpecify null-safety annotations to Redis, Weaviate, Typesense, Qdrant, Pinecone, pgvector, Oracle, OpenSearch, Neo4j, and MongoDB Atlas vector store modules to improve type safety and IDE support #5268
• Added JSpecify null-safety annotations to Milvus and additional vector store modules for improved type safety #5247, #5268
• Made RetryTemplate an optional bean for more flexible configuration 4f121c6
• Updated static analysis suppressions file to reflect current codebase state 1392dda
• Enhanced type safety with JSpecify annotations for Cassandra, Cosmos DB, Azure, and Spring Cloud Bindings modules #5226, #5212, #5213, #5207
• Introduced checkstyle check to ensure all packages include package-info.java documentation 9c734bf
• Reorganized checkstyle suppressions configuration to enable better progress monitoring 0d1ef71
• Enhanced type safety by adding JSpecify annotations to retry module, MCP annotations module, and MCP common module to improve null-safety guarantees. #5207
• Improved code quality by adding JSpecify annotations to CosmosDB, MongoDB, Neo4j, and Redis chat memory repository implementations for better null-safety handling. #5203
• Corrected test classification by moving integration-level test to proper IT test category, improving test organization and build process accuracy. #5208
• Added test coverage for JSpecify nullable field handling in JSON schema generation to ensure proper null-safety annotation support. 316b420
• Resolved issues with the Spring configuration metadata processor. 27a9da0
• Added comprehensive JSpecify annotations across multiple modules including Cassandra and JDBC chat memory repositories, vector store advisors, RAG module, test module, and core vector store module to improve null-safety and type checking. #5203, #5199, #5183
• General updates and improvements to the Neo4j vector store implementation. f3445d1
• Migrated chat-client and vector-store modules to use JSpecify nullability annotations for improved static analysis #5183
• Switched model module to use JSpecify annotations for better null-safety guarantees #5129
• Added JSpecify nullability annotations to newly created subpackages #5164
• Refactored embedding-related files into the model module for better code organization #5159
• Fixed Checkstyle plugin configuration to ensure proper code style enforcement 647a13f
• Improved cache update mechanism with a new implementation approach b104437
• Update CI workflows to use explicit save/restore cache actions for improved build performance and reliability bd56e0b
• Add workflow_dispatch to CI configuration enabling manual workflow execution for testing and troubleshooting #5102
• Make MongoDB starter optional for Testcontainers and Docker Compose modules, reducing required dependencies for non-MongoDB users 08fadd0
• Reworked multiple GitHub Actions workflows into a unified workflow for improved maintainability and consistency #5036
• Resolved GitHub Actions dispatch authorization issue bf5ebce
• Corrected GitHub Actions workflow to ensure commands run in proper git repository context c0246af
• Cleaned up test suite by removing ambiguous test case from ChatClient tools integration test 8909028

🙏 Contributors

Thanks to all the community contributors who made this release possible:

• Arjen Poutsma (@poutsma)
• chabinhwang (@chabinhwang)
• Daniel Garnier-Moiroux (@Kehrlann)
• Eddú Meléndez (@eddumelendez)
• Eric Bottard (@ericbottard)
• guanxu (@guanxuc)
• han-gyeong (@han-gyeong)
• Harrison (@HarrisonC118)
• HartLi (@Hart-Li)
• jiho lee (@destiny3912)
• Joey Surls (@jsurls)
• Julien Dubois (@jdubois)
• JunSeop (@LoperLee)
• Katia Aresti (@karesti)
• Kyle Kreuter (@KyleKreuter)
• lance (@leehaut)
• Manideepok (@Manideepok)
• matejnedic (@MatejNedic)
• Nicolas Krier (@nicolaskrier)
• Pilhwan Choi (@thisishwan2)
• sebin1213 (@sebin1213)
• Thomas Vitale (@ThomasVitale)
• Viacheslav Dobrynin (@viacheslav-dobrynin)
• Wenli Tian (@jamespud)
• Yanming Zhou (@quaff)
• Yokior (@Yokior)
• Yuriy Bezsonov (@ybezsonov)