Bug descriptionZhipu AI model fails to retrieve usage info from ChatResponse upon chat completionEnvironmentSpring AI： 1...

Bug descriptionCalls to DeepSeekChatModel consistently fail with:org.springframework.web.client.RestClientException: Err...

spring ai version: 1.1.0Steps to reproduce TokenTextSplitter splitter = TokenTextSplitter.builder() .withKeepS...

Way back in the start of Spring AI I set the default temperature to 0.7, this was a mistake. The value should not be se...

Prompt and completion data being added to spans (to Observations, as high cardinality data) was previously implemented b...

Bug descriptionWhen using Mistral's Magistral reasoning models (e.g., magistral-medium-2509), the ChatCompletionMessage....

Spring AI 1.1.0-M1spring.ai.mcp.server.protocol=stateless@Componentclass MyTools { @McpTool(description = "say h...

Expected BehaviorI would like Spring AI to support Grounding with Google Maps.It would also be great if the following pa...

spring-ai version: 1.1.0 <dependency> <groupId>org.springframework.ai</groupId> <artif...

Problem StatementThe current architecture implements tool calling as internal recursion within ChatModel implementations...

Spring AI 1.1I have: record UserRandomNumber(int number) { }@McpTool(description = "generate a random number")public ...

Please do a quick search on GitHub issues first, there might be already a duplicate issue for the one you are about to c...

With #4494, VertexAiEmbeddingConnectionAutoConfiguration (and a lot of others) are no longer included in @ImportAutoConf...

When I dynamically configure the OpenAI apikey in the app instead of configuring it in application.yml, app startup erro...

When I use the builder method of ChatModel to build the model, I get a 400 error because I don't know the model paramete...

Bug descriptionWhen LLM returns 400 error message, an empty ChatCompletionChunk was returned from OpenAiApi, which is no...

Bug descriptionChatClientRequest only forbids null keys, not null values. The advisors (ChatModelStreamAdvisor, ChatMode...

Comment From: znxs-dev我也一样，为什么你的OpenAiChatModel类有630行，我追的代码是在380行，发现参数被吞了Comment From: 819318216你是什么版本？1.1.0么

ProblemWhen using ollamaChatModel.call(), the "thinking" metadata field is correctly returned.However, when using the st...

Bug descriptionGetting this error when trying to access the vision capability using ChatClient:java.lang.IllegalStateExc...