Spring AI 1.0.0 RC1 Released

Releases | Mark Pollack | May 13, 2025 | ...

We're excited to announce Spring AI 1.0.0 RC1, marking the final set of breaking changes, bug fixes, and new functionality before the stable release! The GA version is scheduled for May 20th, 2025 - just one week away. During this time, we'll be focusing on improving documentation and addressing any reported bugs.

To celebrate this release, we have added a new song to our AI-generated music playlist Check out the latest track to enhance your blog reading and coding experience.

Important Links

Upgrade Notes
You can automate the upgrade process to 1.0.0-RC1 using an OpenRewrite recipe. This recipe helps apply many of the necessary code changes for this version. Find the recipe and usage instructions at Arconia Spring AI Migrations.

Key Breaking Changes

Chat Client and Advisors

In VectorStoreChatMemoryAdvisor:
- CHAT_MEMORY_RETRIEVE_SIZE_KEY → TOP_K
- DEFAULT_CHAT_MEMORY_RESPONSE_SIZE (100) → DEFAULT_TOP_K (20)
CHAT_MEMORY_CONVERSATION_ID_KEY → CONVERSATION_ID (moved to ChatMemory interface)
- Update imports to: org.springframework.ai.chat.memory.ChatMemory.CONVERSATION_ID

Self-contained Templates in Advisors

Advisors now use independent templates with specific required placeholders:

QuestionAnswerAdvisor: query, question_answer_context
PromptChatMemoryAdvisor: instructions, memory
VectorStoreChatMemoryAdvisor: instructions, long_term_memory

Chat Memory Repository Naming Standardization

In 1.0.0-RC1, we've standardized the naming pattern for chat memory components by adding the repository suffix throughout the codebase. This change affects Cassandra, JDBC, and Neo4j implementations:

Artifact IDs

All memory-related artifacts now follow a consistent pattern:

spring-ai-model-chat-memory-* → spring-ai-model-chat-memory-repository-*
spring-ai-autoconfigure-model-chat-memory-* → spring-ai-autoconfigure-model-chat-memory-repository-*
spring-ai-starter-model-chat-memory-* → spring-ai-starter-model-chat-memory-repository-*

Java Packages

Package paths now include .repository. segment
Example: org.springframework.ai.chat.memory.jdbc → org.springframework.ai.chat.memory.repository.jdbc

Configuration Classes

Main autoconfiguration classes now use the Repository suffix
Example: JdbcChatMemoryAutoConfiguration → JdbcChatMemoryRepositoryAutoConfiguration

Properties

Configuration properties renamed from spring.ai.chat.memory.<storage>... to spring.ai.chat.memory.repository.<storage>...

Observability Changes

Switched from tracing to logging
Properties renamed (e.g., include-prompt → log-prompt)
Removed OTel SDK dependency

All deprecations have been removed for a cleaner API. For complete details, see the Spring AI Upgrade Notes.

New Features

Model Enhancements

DeepSeek Integration

Added dedicated DeepSeek model support with core classes and starter, accommodating its divergence from the OpenAI API.

Azure OpenAI

JSON schema support with builder pattern in AzureOpenAiResponseFormat
Entra ID identity management for simplified authentication
Stream usage support for efficient response handling
reasoning_effort parameter in AzureOpenAiChatOptions

OpenAI

Mutate functionality for OpenAiApi and OpenAiChatModel builders
Web Search Annotations for improved search integration
Configurable imagesPath in OpenAiImageModel

Vertex AI Gemini

Parameter warnings and penalty options for better output quality
JSON array support in jsonToStruct

RAG and Document Processing

DocumentPostProcessors in RAG Advisor for sophisticated transformation pipelines
Customizable text property name with backward compatibility

Tool Calling

Generic argument type support in tool callbacks

Memory Management

Flexible database support in JdbcChatMemory
Cassandra support via CassandraChatMemoryRepository

Observability

Enhanced logging for content observation
Observability support for Bedrock Titan Embedding model

Contributors

There were other refactoring, bug fixing, documentation enhancements across the board by a wide range of contributors. If we haven't gotten to your PR yet, we will, please be patient. Thanks to

Spring blog