Spring AI 1.0.1 Released

Releases | Mark Pollack | August 08, 2025 | ...

On behalf of the Spring AI team and all contributors, I'm happy to announce Spring AI 1.0.1, now available on Maven Central. This point release focuses on stability and bug fixes.

New Song

To celebrate this release, we have added a new song to our AI-generated music playlist. Check out "Fixing the Bugs (Spring AI 1.0.1)" to enhance your blog reading and coding experience.

Release Summary

This release includes 150+ changes focused on stability, enhancements, and documentation improvements.

Community

The Spring AI community continues to grow and contribute in meaningful ways. This release includes contributions from community members who reported issues, submitted fixes, and provided valuable feedback.

Thanks to all those who have contributed with issue reports and pull requests.

Looking Ahead: Spring AI 1.1 and Beyond

While version 1.0.1 focused on stability and bug fixes, the Spring AI team is working on new capabilities for version 1.1. However, with a rapidly evolving AI landscape and a very active PR pipeline to manage, we're being thoughtful about prioritization and would value community input on what matters most.

Our 2025 roadmap diagram provides key dates and shows our planning focus on Spring AI 2.0 with new Spring Boot 4 foundations. The roadmap is primarily date-driven to help the community understand timing, while indicating the major architectural changes we're preparing for the next generation of Spring AI.

Current Focus Areas for Spring AI 1.1

For 1.1, we’re concentrating on a mix of high-impact enhancements and targeted foundational work, with a clear focus on what can realistically land before code freeze. It is still an rather agressive list. We will be reviewing compared against GitHub issues and PRs and look for feedback on what to prioritze.

1. Model Context Protocol (MCP) Support

Deep integration with the latest MCP Java SDK releases, aligning Spring AI with the most up-to-date protocol and transport capabilities:

Multi-protocol version negotiation (2024-11-05 and 2025-03-26).
OAuth2-secured MCP server connections via new transport customizers.
Streamable HTTP and WebMVC/HttpServlet server transports for reactive and servlet deployments.
Structured output validation with JSON Schema enforcement.
Pagination, keep-alive pings, URI template support for richer resource interactions.
Improved error handling, logging, and initialization flows.
Migration to builder-based APIs for tools and transport providers.

2. Core Responses API Enhancements

Expanding the Responses API to close feature gaps, improve provider parity, and bring in the latest SDK capabilities:

Prompt caching to cut latency and cost.
“Thinking” model support for enhanced reasoning.
Message batching for higher throughput.
Native JSON mode and stronger structured output handling across providers.
Hooks for provider-specific extensions while keeping a unified API.
Google Vertex AI SDK update – upgrade to the latest SDK to:
- Unlock newly released endpoints (including non-chat APIs).
- Ensure compatibility with enhanced Responses API features.
- Bring security fixes and long-term support.
- Refresh and expand Vertex AI integration tests.

Special thanks to Dan Dobrin for contributing this work.

3. Chat Memory Improvements

Advancing Spring AI’s memory handling for production scenarios:

Memory compaction to manage token budgets.
Configurable retention policies for long-running conversations.
Improved integration points for custom memory stores.

4. Observability & Multi-Client Configuration

Simplified observability setup, including easier integration with tools like Langfuse.
Multi-client configuration improvements to streamline working with multiple providers in the same application.

5. Net new areas

These are mostly net-new implementations and may extend beyond 1.1 if timing is tight, but early groundwork may begin:

Azure OpenAI – new SDK support.
Vector store improvements, including hybrid search.
Reranking – first-class support for re-ranker models.
Enterprise guardrails – security and compliance features.

6. Items Likely Moving to Incubation

Some larger experiments may be pursued in the community before core inclusion:

MemGPT-style chat memory implementations.
AgentClient for running autonomous CLI agents (e.g. Claude Code) via Spring AI.

We’ll continue to adjust priorities as we work toward the September 23 code freeze for 1.1, balancing near-term deliverables with strategic groundwork for Spring AI 2.0.

Community Guidance Needed:

We can't tackle everything simultaneously, so community feedback on prioritization is invaluable. Tell us what to prioritize next in Issues or via PRs.

The team is also investing in AI-powered tooling to help manage our growing backlog more efficiently, but your input remains essential for steering the project's direction.

🙏 Contributors

Thanks to all contributors who made this release possible:

Resources

Project Page | GitHub | Issues | Documentation | Stack Overflow

Spring blog