LaunchedEditorial Listing

Microsoft Agent Framework

Microsoft · Multi-Agent Orchestration Framework

Microsoft's production-ready framework for building, orchestrating, and deploying AI agents and multi-agent workflows in Python and .NET (with Go in preview) — the successor to both AutoGen and Semantic Kernel.

PricingFree

Setupmedium

Runs onSelf-hosted · Web

APIYes

Open sourceYes

DocsYes

CategoryAgent Frameworks

Agent FrameworkPython.NETGoOpen SourceMicrosoftMulti-Agent

Best for

Python and .NET multi-agent orchestration, especially with Azure and Microsoft Foundry

Not ideal for

Non-developers or teams outside the Microsoft ecosystem

Who it's for

Python and .NET developers building production multi-agent systems, especially within the Microsoft ecosystem

Capabilities

Python and .NET SDKs (GA), plus a Go SDK in public preview
Graph-based multi-agent workflows with checkpointing, human-in-the-loop, and time-travel
Middleware, OpenTelemetry observability, and DevUI
Declarative YAML agents and Agent Skills
MCP and A2A interoperability
Microsoft Foundry integration, including Foundry Hosted Agents
Works with OpenAI, Anthropic, Amazon Bedrock, Gemini, Ollama, and more

Limitations

Requires developer expertise in Python or .NET
Strongest within the Microsoft and Azure ecosystem
Self-managed by default; managed hosting requires Microsoft Foundry (Azure)
Fast release cadence (frequent minor versions); Go SDK still in preview

Use cases

Building multi-agent workflows in Python or .NET
Enterprise AI agent systems with Azure and Foundry integration
Migrating from AutoGen or Semantic Kernel
Graph-based agent orchestration with checkpoints and human approval
Conversational AI agents with complex workflows

Integrations & fit

Microsoft FoundryAzure OpenAIOpenAIAnthropicAmazon BedrockGeminiOllamaGitHub Copilot SDKMCPA2A

Good fit forStartup / small team, Enterprise

Pricing modelFree· No cost to start

See pricing on Microsoft Agent Framework →

Alternatives to consider

LangGraph

LangChain

Developers building production multi-agent systems that need fine-grained control over state, execution flow, and human-in-the-loop checkpoints — and who are willing to trade setup time for that control

Free

CrewAI

Orchestrating autonomous agent teams for enterprise tasks

Freemium

AutoGen

Microsoft

Existing AutoGen users maintaining projects, and developers experimenting with conversational multi-agent patterns or building iterative workflows (code generation + review, research + verification) where the solution emerges from agent dialogue rather than a predefined execution graph

Free

Google ADK

Google

Developers building production agents in Python, TypeScript, Go, Java, or Kotlin who want an open-source, code-first framework with first-class Gemini support and a direct path to Google Cloud

Free

About Microsoft Agent Framework

Microsoft Agent Framework (MAF) is an open-source SDK for building AI agents and multi-agent workflows. It is Microsoft's direct successor to both AutoGen (now in maintenance mode) and Semantic Kernel, built by the same teams, and it shipped a production-ready 1.0 release with stable APIs and a long-term support commitment; migration guides exist for both predecessors. It supports Python and .NET, with a Go SDK in public preview, and offers graph-based workflows with checkpointing, human-in-the-loop steps, and time-travel, plus middleware, OpenTelemetry observability, declarative YAML agents, a DevUI for local testing, Agent Skills, and MCP and A2A interoperability. The framework integrates closely with Microsoft Foundry and Azure but also works with other providers such as OpenAI, Anthropic, Amazon Bedrock, Gemini, and Ollama. Agents run on your own infrastructure by default, or can be deployed to Microsoft-hosted infrastructure with Foundry Hosted Agents. The SDK is MIT-licensed and available via pip (Python) and NuGet (.NET), with documentation on Microsoft Learn.

Are you the founder? Claim this listing →