15 Pull requests merged by 13 people

• docs: add description of embedding-field-name property in elasticsearch.adoc

  #3399 merged May 30, 2025

• Add unit tests for OpenAiStreamFunctionCallingHelper

  #3312 merged May 30, 2025

• Avoid looping through all the page tree if there is no destination page

  #3396 merged May 30, 2025

• fix: Remove unreachable else-if branch in PromptChatMemoryAdvisor

  #3395 merged May 30, 2025

• feat(anthropic): add support for claude-opus-4 and claude-sonnet-4

  #3351 merged May 30, 2025

• Fix CustomerTools example to use existing methods

  #3390 merged May 30, 2025

• Bump org.apache.pdfbox to 3.0.4 and guard against empty unicode strings

  #3271 merged May 30, 2025

• docs: fix typo 'synch' to 'sync' in CONTRIBUTING.adoc

  #3378 merged May 29, 2025

• Adding checkstyle notes in README and contribution guidelines

  #3277 merged May 29, 2025

• Update mistralai-embeddings.adoc

  #3362 merged May 28, 2025

• fix typo in ToolCallback

  #3352 merged May 27, 2025

• Fix links in documentation to Media.java

  #3322 merged May 27, 2025

• Add metadata support for vector store types

  #3325 merged May 27, 2025

• Fix awesome-spring-ai link to official repository

  #3328 merged May 27, 2025

• docs: update custom API key example using OpenAiChatModel.builder()

  #3335 merged May 27, 2025

19 Pull requests opened by 15 people

• fix: omit temperature for search models and add tests

  #3315 opened May 24, 2025

• Fix MCP protocol, related commands cannot be executed under Windows, such as npx

  #3323 opened May 26, 2025

• Refactor builder pattern in VertexAiGeminiSafetySetting

  #3326 opened May 26, 2025

• Remove redundant validation call in getToolCallbacks method

  #3329 opened May 26, 2025

• support for similarity threshold in VectorStoreChatMemoryAdvisor

  #3341 opened May 27, 2025

• feat: Add tool_calls support to JdbcChatMemoryRepository

  #3343 opened May 27, 2025

• fix(ai-rag): Fix query expansion result handling logic, same as #3345

  #3347 opened May 27, 2025

• feat(anthropic): Implemented a web search tool, provided by Anthropic

  #3353 opened May 27, 2025

• fix issue@ #2399

  #3356 opened May 28, 2025

• Add support for ApiKey for Anthropic and use it dynamically for every request

  #3365 opened May 28, 2025

• fix(qdrant): support Long type in payload by converting to String

  #3367 opened May 29, 2025

• Update OpenAiChatModel.java

  #3376 opened May 29, 2025

• fix: OpenAiChatOptions support transfer webSearchOptions

  #3377 opened May 29, 2025

• ToolCallingChatOptions support internalToolExecutionMaxAttempts, to limit the maximum number of tool calls and prevent infinite recursive calls to LLM in special cases

  #3380 opened May 29, 2025

• Upgrade Spring Boot to 3.5.0

  #3381 opened May 29, 2025

• Added support for the "think" for Ollama

  #3386 opened May 30, 2025

• Run PR Check GHA on every branch and allow it in Forks

  #3391 opened May 30, 2025

• Create new ToolChatOptions if input request has tool configuration without options

  #3397 opened May 30, 2025

• Resolve OpenAI ApiKey for every request as late as possible

  #3398 opened May 30, 2025

19 Issues closed by 11 people

• ElasticsearchVectorStore

  #2082 closed May 30, 2025

• JdbcChatMemoryRepositoryProperties注入失败

  #3360 closed May 30, 2025

• Streaming output is not token-by-token when using tools in ChatClient prompt (Ollama Qwen3:8b)

  #3373 closed May 30, 2025

• ollamaModel stream+tools : NullPointerException "evalDuration" is null

  #3369 closed May 30, 2025

• Timeline for Spring AI GA Release

  #1273 closed May 30, 2025

• ElasticsearchVectorStore create error

  #3387 closed May 30, 2025

• Vald

  #1338 closed May 30, 2025

• [Vector store] Clickhouse support

  #3249 closed May 30, 2025

• 1.0.0 not has ChatClient

  #3384 closed May 30, 2025

• Exception reading PDF files in PagePdfDocumentReader

  #3054 closed May 30, 2025

• java.lang.NoSuchMethodError: 'void org.apache.tika.parser.pdf.PDF2XHTML.setIgnoreContentStreamSpaceGlyphs(boolean)'

  #3265 closed May 30, 2025

• RAG documentation doesn't mention spring-ai-rag dependency

  #3374 closed May 29, 2025

• Help me, my mcp tool doesn't work。I would be very grateful if anyone could help me。

  #3291 closed May 29, 2025

• About Function Calling

  #3316 closed May 28, 2025

• MCP Server java.lang.OutOfMemoryError when handling multiple MCP client connections or reconnections

  #3319 closed May 27, 2025

• RedisVectorStoreAutoConfiguration config failed with multi embeddingmodels

  #3346 closed May 27, 2025

• When using Chat Memory, if it is in streaming mode, an error "No StreamAdvisors available to execute" will occur. In non - streaming mode, it runs normally.

  #3344 closed May 27, 2025

• No schema scripts found at location 'classpath:org/springframework/ai/chat/memory/repository/jdbc/schema-mysql.sql'

  #3336 closed May 27, 2025

• When use @Autowire List<McpSyncClient>mcpSyncClients, If one of the configured MCP servers is invalid, it will cause the system to fail to start!

  #3306 closed May 24, 2025

31 Issues opened by 29 people

• Will `returnDirect = true` effect in mcp server?

  #3400 opened May 30, 2025

• Add stream_options Record to ChatCompletionRequest in Spring AI DeepSeek Provider

  #3394 opened May 30, 2025

• Adding custom CardinalityKeyValue did not work

  #3393 opened May 30, 2025

• configuring defaultOptions can cause defaultToolCallbacks failure

  #3392 opened May 30, 2025

• ToolCallback fails if tool context is provided

  #3389 opened May 30, 2025

• AssistantMessage supports reasoningContent

  #3388 opened May 30, 2025

• Two types of vectordatabase are imported into the project pom.xml. When starting the project, an error is reported, indicating that vectorstore has been defined.

  #3385 opened May 30, 2025

• Accessing the New "Thought" Field from Ollama in the Spring AI Framework

  #3383 opened May 30, 2025

• [EPIC] Spring Boot 4.0 compatability

  #3379 opened May 29, 2025

• anthropic-ai thinking error

  #3371 opened May 29, 2025

• "Request was interrupted: null" error when VectorStore.save is called with a List<Document> whose size is large

  #3370 opened May 29, 2025

• Migrate from Azure OpenAI Java SDK to OpenAI Java Library

  #3368 opened May 29, 2025

• Streaming mode AssistantMessage does not retain toolCalls, causing issues with tool confirmation workflows

  #3366 opened May 29, 2025

• How to set metadata data information when calling MCP?

  #3363 opened May 28, 2025

• You should not need to provide an openai api key in application.properties if you set it manually via code every time you call it

  #3361 opened May 28, 2025

• tool-response-mime-type property conflicts with ToolCallResultConverter class

  #3358 opened May 28, 2025

• The isFunctionalType method of MethodToolVNet Provider is incorrect, resulting in some cases where the tool cannot be registered

  #3355 opened May 28, 2025

• Azure responseFormat no converter found

  #3354 opened May 27, 2025

• When using `MultiQueryExpander`, if the model's response contains empty newlines, the query expansion fails and returns the original query unchanged. The current implementation doesn't properly handle or filter out empty lines from the model's response.

  #3345 opened May 27, 2025

• Support persisting tool_calls in JdbcChatMemoryRepository

  #3342 opened May 27, 2025

• ToolResponseMessage results in null content, causing SQLIntegrityConstraintViolationException in JdbcChatMemoryRepository

  #3339 opened May 27, 2025

• Why is `ToolMetadata` not designed as extensible, currently only has one `returnDirect` attribute?

  #3338 opened May 27, 2025

• After the mcp server restarts, the client will return 404.

  #3337 opened May 27, 2025

• Max tool calls / Max internalCall limit

  #3333 opened May 26, 2025

• 用SpringAI实现一个MCP Server，怎么接收客户端传递过来的env

  #3330 opened May 26, 2025

• Application does not start when using 2GB+ models

  #3327 opened May 26, 2025

• Hope to support Whisper in the future

  #3324 opened May 26, 2025

• Mongo vectorStore.similaritySearch failing due to org.bson.types.ObjectId cannot be cast to class java.lang.String

  #3318 opened May 24, 2025

• Pluggable Java-method tool-definition strategies & typed inputSchema in MCP

  #3317 opened May 24, 2025

• ZhiPuAi and MiniMax ChatModel support completions-path properties config

  #3314 opened May 24, 2025

• Vertex AI Streaming not working

  #3313 opened May 24, 2025

36 Unresolved conversations

Sometimes conversations happen on old items that aren’t yet closed. Here is a list of all the Issues and Pull Requests with unresolved conversations.

• Escape asterisk

  #2722 commented on May 28, 2025 • 2 new comments

• Add OpenAI's Image edit feature

  #2914 commented on May 26, 2025 • 1 new comment

• Create MongoDB chat memory implementation

  #2679 commented on May 30, 2025 • 0 new comments

• Add filter query support for GemFireVectorStore

  #2753 commented on May 29, 2025 • 0 new comments

• feat: options and ChatCompletionRequest add property enable_thinking

  #2940 commented on May 30, 2025 • 0 new comments

• Support pre-parsed Filter.Expression in getFilterExpression

  #3068 commented on May 29, 2025 • 0 new comments

• add aws bedrock prompt-caching support

  #3213 commented on May 30, 2025 • 0 new comments

• feat(rag:etl): Add custom template support to KeywordMetadataEnricher

  #3252 commented on May 30, 2025 • 0 new comments

• Fix NPE caused by uninitialized static variables in interfaces in native images

  #3256 commented on May 30, 2025 • 0 new comments

• Provides a Jinja2 implementation for TemplateRenderer

  #3259 commented on May 29, 2025 • 0 new comments

• Add Spring SpEL implementation of TemplateRenderer

  #3262 commented on May 29, 2025 • 0 new comments

• Fixed a bug in the augmentSystemMessage method of the Prompt.

  #3268 commented on May 29, 2025 • 0 new comments

• fix: correct visibility scope issues in Spring AI components

  #3284 commented on May 29, 2025 • 0 new comments

• Add tests for LastMaxTokenSizeContentPurger

  #3293 commented on May 29, 2025 • 0 new comments

• fix(deepseek): prevent incorrect ToolCall merging caused by empty id and name strings

  #3298 commented on May 30, 2025 • 0 new comments

• Fix PropmptTemplate typo in structured-output-api.jpg

  #3304 commented on May 29, 2025 • 0 new comments

• Add support for max_completion_tokens in AzureOpenAI chat options request and update the test and document.

  #3305 commented on May 29, 2025 • 0 new comments

• build: Use spring-web instead of spring-webflux in spring-ai-retry

  #3308 commented on May 29, 2025 • 0 new comments

• MCP integration using WebMVC SSE

  #3287 commented on May 25, 2025 • 0 new comments

• Support A2A Protocol

  #2911 commented on May 25, 2025 • 0 new comments

• this is a bug for spring-ai-mcp-client of deepseek-v3 by flux type

  #3297 commented on May 26, 2025 • 0 new comments

• ChatClient returns 401 Invalid API Key when using Groq as OpenAI-compatible provider

  #3289 commented on May 27, 2025 • 0 new comments

• 使用spring-ai-starter-mcp-client sse方式连接mcp server，重启mcp server后，以前的client不会自动重连。

  #2740 commented on May 27, 2025 • 0 new comments

• Failed to send message to session xxx: Response not usable after response errors.

  #3061 commented on May 27, 2025 • 0 new comments

• Output of reasoning_content that supports deepseek-r1 is required

  #2283 commented on May 27, 2025 • 0 new comments

• Support for Text Chunking with Overlap in TokenTextSplitter

  #2123 commented on May 27, 2025 • 0 new comments

• Agent support

  #403 commented on May 28, 2025 • 0 new comments

• Support for OpenAI Responses API

  #2962 commented on May 28, 2025 • 0 new comments

• Support for Gemini 2.5?

  #3303 commented on May 28, 2025 • 0 new comments

• 当需要调用MCP Server的多个tool时，MCP Client在调用第二个tool时出现反序列化错误（When multiple tools of the MCP Server need to be called, the MCP Client encounters a deserialization error when calling the second tool.）

  #2818 commented on May 29, 2025 • 0 new comments

• RetrievalAugmentationAdvisor appears to suppress tool usage

  #3310 commented on May 29, 2025 • 0 new comments

• Model incompatible request argument supplied: temperature

  #3253 commented on May 29, 2025 • 0 new comments

• io.grpc.StatusRuntimeException: INTERNAL: Encountered end-of-stream mid-frame

  #2594 commented on May 30, 2025 • 0 new comments

• RedisVectorStore search incorrect

  #2514 commented on May 30, 2025 • 0 new comments

• Can multiple MCP servers with different endpoints be published in one service?

  #3299 commented on May 30, 2025 • 0 new comments

• Support accessing HTTP headers in ToolContext when using WebFlux

  #2757 commented on May 30, 2025 • 0 new comments