# Spring AI 2.1.0-RC2 Available Now

DevFeed: [Spring AI 2.1.0-RC2 Available Now](<https://devfeed.tech/articles/spring-ai-2-1-0-rc2-available-now-122954.md>)

Original publisher: [Read original article](<https://spring.io/blog/2026/11/09/spring-ai-2-1-0-RC2-available-now>)

Author: dimitarproynov

Published: 2026-11-09T00:00:00Z

Content type: release

Language: en

Sources: [Spring](<https://devfeed.tech/sources/spring.md>)

Topics: [Spring AI](<https://devfeed.tech/topics/spring-ai.md>), [AI Chat](<https://devfeed.tech/topics/ai-chat.md>), [MSP MCP](<https://devfeed.tech/topics/msp-mcp.md>), [Streaming](<https://devfeed.tech/topics/streaming.md>), [Embeddings](<https://devfeed.tech/topics/embeddings.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [anthropic](<https://devfeed.tech/tags/anthropic.md>), [api](<https://devfeed.tech/tags/api.md>), [apis](<https://devfeed.tech/tags/apis.md>), [chat](<https://devfeed.tech/tags/chat.md>), [choice](<https://devfeed.tech/tags/choice.md>), [decision-models](<https://devfeed.tech/tags/decision-models.md>), [decisionmodel](<https://devfeed.tech/tags/decisionmodel.md>), [delta](<https://devfeed.tech/tags/delta.md>), [embeddings](<https://devfeed.tech/tags/embeddings.md>), [java-sdk](<https://devfeed.tech/tags/java-sdk.md>), [maven](<https://devfeed.tech/tags/maven.md>), [maven-central](<https://devfeed.tech/tags/maven-central.md>), [mcp](<https://devfeed.tech/tags/mcp.md>), [mcp-protocol](<https://devfeed.tech/tags/mcp-protocol.md>), [probability](<https://devfeed.tech/tags/probability.md>), [score](<https://devfeed.tech/tags/score.md>), [server](<https://devfeed.tech/tags/server.md>), [spring-ai](<https://devfeed.tech/tags/spring-ai.md>), [streaming](<https://devfeed.tech/tags/streaming.md>)

## AI overview

Spring AI 2.1.0-RC2 adds `DecisionModel` for typed yes/no, choice, and score responses with probabilities, letting applications set confidence thresholds; answers remain nondeterministic. Implementations include OpenAI, Ollama, and `TypeSafeClient`. New reranking and filtering features use decision calls, run per document over an executor, and fail open by default; the filter can fail closed. Tool argument deltas are exposed as partial `ToolCallPart`s for Anthropic, OpenAI, Bedrock Converse, and DeepSeek, while only the completed call is actionable or aggregated. User-supplied embedding upsert and search expand to five additional vector stores, though the excerpt does not name them. The MCP server adds protocol revision 2026-07-28 (`MODERN`), with synchronous support only; async mode fails at startup, and some existing context functions have no modern equivalent yet. Upgrade changes include removing deprecated Anthropic sampling settings rejected by its newer SDK, plus adjustments to streaming, Typesense, and token-usage behavior.

## Source excerpt

On behalf of the Spring AI team and everyone who has contributed, I'm happy to announce that Spring AI 2.1.0-RC2 has been released and is now available from Maven Central! Release notes | Upgrade notes | Reference documentation 2.1.0-RC2 builds on 2.1.0-RC1 and delivers DecisionModel, a new API for typed yes/no, choice and score decisions, together with the first things that are built on it. It also adds tool call delta streaming, extends the pre-computed embeddings support to five more vector stores, brings server support for the new MCP protocol revision, and upgrades to the Anthropic Java SDK 3.x. New in this release Decision Models Many decisions in an application have a small, known set of possible results: yes or no, one of a few categories, or a level on a scale. You could ask a chat model, but it answers with free text, and you must turn that text into a reliable result yourself. The new DecisionModel API is made for these decisions. You describe the possible results, and the model picks one and tells you how likely each result is, instead of generating text. It follows the shape of the other model APIs: a DecisionPrompt goes in, a DecisionResponse comes out, and DecisionOptions tune the call. A prompt holds an input (a text, a set of named fields, or a conversation) and one or more named questions: YesNoQuestion, answered with the probability of yes. ChoiceQuestion, answered with the selected option and, when the provider reports them, the probability of each option. ScoreQuestion, answered with a level of an ordered scale, such as unhappy, neutral, happy. DecisionResponse response = decisionModel.call(DecisionPrompt.builder() .text("My new headphones arrived yesterday, but the left side doesn't work. Can I get my money back?") .question("defect", YesNoQuestion.builder() .text("Does the customer report a defect?") .whenYes("Part of the item does not work as it should") .whenNo("The item works, and the message is about something else") .build()) .question("t