Semantic Kernel
The latest news from the Semantic Kernel team for developers
Latest posts
Integration of AWS Bedrock Agents in Semantic Kernel
Overview of AWS Bedrock Agents AWS Bedrock Agents provide a managed service that facilitates the experimentation and rapid deployment of AI agents. Users can leverage proprietary AWS models as well as a diverse selection of models from various providers available on AWS Bedrock. Semantic Kernel's Integration with AWS Bedrock Semantic Kernel now integrates with AWS Bedrock Agents, enabling users to leverage kernel functions alongside features such as code interpretation and Retrieval-Augmented Generation (RAG) powered by the AWS knowledge base. If you manage resources on AWS and are exploring a multi-cloud AI s...
Talk to your agents! Introducing the Realtime API’s in Semantic Kernel
Introducing Realtime Agents in Semantic Kernel for Python With release 1.23.0 of the Python version of Semantic Kernel we are introducing a new set of clients for interacting with the realtime multi-modal API's of OpenAI and Azure OpenAI. They provide a abstracted approach to connecting to those services, adding your tools and running apps that leverage these very powerful and useful agents. The key addition that Semantic Kernel brings when you want to connect to these models is that we make the experience of using these models with function calling very easy, just create a Kernel and add your plugins as ...
Effortlessly Integrate xAI’s Grok with Semantic Kernel
For Semantic Kernel users, integrating xAI's Grok API using the OpenAI connector is a breeze thanks to its compatibility with OpenAI's API format. This tutorial focuses on setting up Grok in your Semantic Kernel projects with minimal fuss, using C# and Python examples. Why Grok? Grok, built by xAI, is a powerful AI model, offers a 128k context window and function-calling support, making it a solid choice for complex tasks in Semantic Kernel. With an API compatible with OpenAI, announced in November 2023 and now available via API access, with models like "grok-beta" available for developers and soon the ne...
AutoGen and Semantic Kernel, Part 2
Following on from our blog post a couple months ago: Microsoft’s Agentic AI Frameworks: AutoGen and Semantic Kernel, Microsoft’s agentic AI story is evolving at a steady pace. Both Azure AI Foundry’s Semantic Kernel and AI Frontier’s AutoGen are designed to empower developers to build advanced multi-agent systems. The AI Frontier’s team is charging ahead pushing the boundaries of multi-agent approaches, building new agentic patterns as well as a growing library of purpose-built agents, such as Magentic One, while the Semantic Kernel team builds on years of enterprise expertise to enable developers to build agents...
Integrating Model Context Protocol Tools with Semantic Kernel: A Step-by-Step Guide
This post describes how to use Model Context Protocol tools with Semantic Kernel. Model Context Protocol (MCP) is an open protocol that standardizes how applications provide context to LLMs. MCP standardizes the connection between AI models and various data sources and tools. The Model Context Protocol is significant because it enhances the way AI models interface with data and tools, promoting interoperability, flexibility, and improved contextual understanding. Its potential applications span various domains including, data integration and knowledge management, making it a valuable component in the developme...
Guest Blog: LLMAgentOps Toolkit for Semantic Kernel
Today the Semantic Kernel team is excited to welcome a guest author, Prabal Deb to share his work. LLMAgentOps Toolkit is repository that contains basic structure of LLM Agent based application built on top of the Semantic Kernel Python version. The toolkit is designed to be a starting point for data scientists and developers for experimentation to evaluation and finally deploy to production their own LLM Agent based applications. Architecture The LLMAgentOps architecture might be constructed using the following key components divided into two phases like DevOps / MLOps / LLMOps development and deploym...
Release the Agents! SK Agents Framework RC1
Semantic Kernel Agent Framework Reaches Release Candidate 1 We're excited to announce that with the release of Semantic Kernel 1.40 (.NET) and 1.22.0 (Python), we're elevating the Semantic Kernel Agent Framework to Release Candidate 1. This marks a significant milestone in our journey toward providing a robust, versatile framework for building AI agents for enterprise applications. Code Sample: Creating a Chat Agent with tool plugins Creating an agent with Semantic Kernel is easy! Let's look at a simple Python example that demonstrates how to create a chat completion agent that answers questions about a menu u...
Compatibility of PostgreSQL Connector with AWS and GCP
As AI-driven applications continue to evolve, the need for efficient vector-based search capabilities is greater than ever. Microsoft Semantic Kernel makes it easy to integrate these capabilities with PostgreSQL databases using the connector. Whether you're leveraging cloud-hosted PostgreSQL instances on Amazon Web Services or Google Cloud, this connector enables seamless interaction, allowing you to store and query vectorized data for tasks like recommendation systems, semantic search, and more. Compatible Databases Semantic Kernel Postgres Connector is compatible with PostgreSQL instances hosted locally or i...
Hybrid Model Orchestration
Hybrid model orchestration is a powerful technique that AI applications can use to intelligently select and switch between multiple models based on various criteria, all while being transparent to the calling code. This technique not only allows for model selection based on factors such as the prompt's input token size and each model's min/max token capacity, or data sensitivity - where sensitive inference is done against local models and the others against cloud models - returning either the fastest response, the most relevant response, or the first available model's response, but also provides a robust fallbac...