Edit

Build AI apps and agents with orchestration frameworks for Azure HorizonDB (Preview)

Azure HorizonDB integrates with leading LLM orchestration frameworks for building both AI applications and autonomous agents. These frameworks provide the scaffolding for managing LLMs, embedding models, agent memory, multi-agent workflows, and database connectivity.

Framework integrations

The following frameworks and services integrate with Azure HorizonDB:

Framework Description Azure HorizonDB integration
Microsoft Agent Framework (incorporating Semantic Kernel) A unified, production-ready open-source SDK by Microsoft for building AI agents in .NET and Python. It merges the orchestration and plugin capabilities of Semantic Kernel with multi-agent workflow patterns. Python connector
.NET connector
Java connector
LangChain / LangGraph LangChain simplifies LLM application development with tools for context-aware reasoning in Python and JavaScript. LangGraph adds stateful, graph-based orchestration for complex multi-agent workflows with checkpointed execution. Develop AI apps with LangChain and Azure HorizonDB (Preview)
JavaScript
LlamaIndex A framework for building context-augmented AI applications that integrates private or domain-specific data with LLMs for complex workflows. Excels at structured data retrieval and knowledge-graph-powered reasoning. Python
GraphRAG A framework by Microsoft that uses Azure HorizonDB to create AI-powered knowledge graphs. It enables robust data models and reveals relationships in semistructured data. Quickstart
CrewAI An open-source framework for orchestrating role-based, collaborative multi-agent workflows with task delegation and standard operating procedures. PGSearchTool
Auto-Gen A Microsoft framework for multi-agent conversation patterns. Supports flexible agent communication topologies and tool integration. PostgreSQL tools
Microsoft Foundry Agent Service A managed service for building, deploying, and scaling AI agents with built-in tool support, tracing, and monitoring. Overview