As we move into the first half of 2025, I’m excited to share our ambitious roadmap that we hope will enable you to build even more sophisticated AI applications with Semantic Kernel.
Agent Framework 1.0
By the end of Q1 2025, the SK Agent Framework will transition from preview to general availability (GA). This milestone marks a commitment to a stable, versioned API that minimizes breaking changes and supports production-grade applications. As highlighted in our recent community office hours meeting, our vision is to embrace an agent-first programming model. This approach has been shaped by extensive feedback from our developer community and has been informed by our long-standing experience with building agentic applications. With this GA release, you can expect:
-
Enhanced Stability: We’re moving the API out of preview on into general availability with v1.0. We will add non-breaking updates as the ecosystem evolves, but we recognize the importance of a stable and supported base platform.
-
Integration with AutoGen: A major highlight of our agents roadmap is the strategic convergence between Semantic Kernel and AutoGen. This integration is unfolding in three ways:
-
AutoGen integrates with Semantic Kernel: AutoGen now leverages the powerful capabilities of Semantic Kernel [link: https://devblogs.microsoft.com/semantic-kernel/microsofts-agentic-ai-frameworks-autogen-and-semantic-kernel/], allowing developers to integrate in a familiar way.
-
Hosting AutoGen agents in Semantic Kernel: You’ll soon have the flexibility to host AutoGen agents within the Semantic Kernel ecosystem. This means that if you’re already working with AutoGen, you can seamlessly migrate and manage your agents in Semantic Kernel.
-
Converging Agent Runtimes: We are actively working on harmonizing the core components between AutoGen and Semantic Kernel. This convergence will simplify development workflows and enhance the interoperability of multi-agent systems. We’re planning a blog post with more technical details on this in the coming week.
-
-
Support for other agent frameworks: In addition to AutoGen, last week, Semantic Kernel added support for Azure AI Foundry Agents Service. This allows you to build agentic and multi-agent systems in Semantic Kernel that orchestrate agents deployed to Azure AI Foundry. Over the coming weeks, we also plan to add support for other popular agent frameworks.
Simplifying Business Workflow Orchestration with the Semantic Kernel Process Framework
In parallel with our Agent Framework enhancements, we are continuing work on the Semantic Kernel Process Framework. We’re planning to release this framework out of preview by the end Q2 2025. As we put the finishing touches on the Process Framework, now is the right time to give a try with your business workflows. We’ve added a substantial number of new features in samples for both C# and Python.
Tooling and Visualization
In the next few weeks, we’re going to refresh our Visual Studio Code extension to complement the new capabilities we’ve added to Semantic Kernel over the last few months. We’re planning to add support for our unified Semantic Kernel declarative format, and the ability to visualize and deploy agent and process workflows from within VS Code.
Other Major Investments
In addition to the core areas mentioned above, we are making significant investments to broaden the Semantic Kernel ecosystem:
- Expanded Connectors and Models: We’re continuously working to integrate more connectors, support additional models, and enhance agent functionalities to meet a wide range of application scenarios, highlighted by our recent DeepSeek update.
- OpenAI Realtime Audio API: This integration will open new avenues for real-time voice and audio-enabled AI applications.
- Memory Enhancements: Improvements in memory abstraction will facilitate more sophisticated and context-aware agent behaviors.
Looking Ahead
The next six months promise to be a period of substantial innovation and growth for Semantic Kernel. Our focus on agents, process orchestration, and enhanced cloud integration is a direct response to your feedback and the evolving landscape of AI application development. We built Semantic Kernel for you, and we invite you to experiment with these new features, share your feedback, and contribute to the ongoing evolution of Semantic Kernel. Together, we’re building the future of enterprise AI applications.
For any questions or to provide feedback, please join the discussion on GitHub repository, Discord channel and our community calls.
Happy coding!
0 comments
Be the first to start the discussion.