AutoGen v0.4 vs. AG2: The Microsoft Community Split Explained

AutoGen vs. AG2 refers to the split of the original AutoGen project into two lines: Microsoft rebuilt AutoGen with version v0.4 on an event-driven, actor-based architecture, while part of the community continued the classic, conversation-based code under the name AG2. Both lines share the basic concept of conversation-based multi-agent systems, but differ in origin, governance, architecture and roadmap. For DACH decision-makers, the central question is not "which fork is better" but "which path is future-proof for my project".

• AutoGen v0.4 is Microsoft's re-architecture: event-driven, actor-based, designed for scaling and distributed agents.
• AG2 is the community fork that continues the familiar, conversation-based API of the original AutoGen line (as of v0.2).
• Strategically, Microsoft consolidates its agent roadmap in the Microsoft Agent Framework 1.0 (GA since 3 April 2026, MIT licence), which brings together Semantic Kernel and AutoGen — meaning AutoGen is considered to be in maintenance mode.

Note on context: The robustly verified facts in this article concern the Microsoft Agent Framework 1.0 (GA date, licence, providers, EU regions) as well as the AutoGen maintenance status according to Microsoft's roadmap. The details on AutoGen v0.4 and the AG2 fork are based on publicly documented project knowledge of the respective repositories and should be cross-checked directly there before any production decisions (as of 2026).

Origin and Governance: How the Split Came About

AutoGen originally emerged as a research and open-source project from the Microsoft environment and quickly became one of the best-known frameworks for multi-agent conversations. The original line (around version v0.2) popularised the framework through so-called conversable agents: agents that "talk" to each other in natural language, negotiate tasks and call tools.

With the further development to v0.4, Microsoft opted for a fundamental architectural rebuild rather than an evolutionary continuation. This break — new architecture, changed API philosophy — was the trigger for the community split. Part of the original contributors continued the proven, conversation-centric codebase under the name AG2. This created two parallel lines:

• AutoGen (Microsoft): governance at Microsoft, strategically embedded in the company's larger agent portfolio.
• AG2 (community): community-driven further development with a focus on continuity for existing users of the classic API.

Important for context: Microsoft consolidated its agent strategy in 2026. The Microsoft Agent Framework 1.0 has been generally available since 3 April 2026 (public preview since 1 October 2025) and unites Semantic Kernel and AutoGen under the MIT licence. Before the merger, the predecessor projects together accumulated over 75,000 GitHub stars. AutoGen concepts are integrated into the new graph workflow of the Agent Framework. According to Microsoft's roadmap — this statement circulates primarily via secondary sources and should be cross-checked against the official AutoGen project page — AutoGen is in maintenance mode, while the strategic investment flows into the Agent Framework.

Architecture Differences: Conversation vs. Actors

The technical core of the split lies in the architecture model.

AutoGen v0.4 pursues an event-driven, actor-based approach. Instead of a linear conversation flow, agents communicate via messages and events. This resembles classic actor models from distributed programming: each agent is a self-contained unit that reacts to incoming messages. The advantages are better scalability, asynchronicity and suitability for distributed systems. The price is higher conceptual complexity compared to the simple "two agents talk to each other" model.

AG2 sticks to the conversation-centric model of the original line. The central building blocks are conversable agents that collaborate in group chats and via handoffs. This model is more intuitive for beginners and sufficient for many prototyping scenarios — the mental load is lower, and the API is more familiar to anyone who has already worked with classic AutoGen.

This distinction is more than cosmetic. It determines how you structure multi-agent logic, how you debug and how well the system scales under load.

Table: AutoGen v0.4 vs. AG2 Compared

Note: Specific version states and star counts of both projects are snapshots and should be checked directly in the respective repository before any production decisions (as of 2026).

Conversation Agents: What Stays, What Changes

Both lines descend from the same conceptual root: agents that coordinate via natural language, call tools and divide tasks among several roles. In AG2 this pattern is the main purpose — an assistant agent, a user proxy agent and a group chat manager are enough to set up a collaborative pipeline.

In AutoGen v0.4 the same functional goal exists, but is realised via the event layer: conversations are a pattern on top of the message backbone, not the foundation itself. In the Microsoft Agent Framework, this line ultimately merges into the graph workflow, which models orchestration patterns (sequential, parallel, hierarchical, concurrent) explicitly as a graph — i.e. more structured and closer to production than free-form conversation.

Concrete Example: Decision for a DACH SME Project

A Vienna-based B2B software company wants to build a multi-agent process for automated proposal generation: a research agent, a calculation agent, a review agent. The team faces the choice.

• Existing PoC in classic AutoGen (v0.2): The existing prototype runs with conversable agents. If it is to live on without an architecture break, AG2 is the pragmatic path — the API is almost identical, and the migration effort is minimal.
• Strategic rebuild within the Microsoft stack: If the company already uses Azure and needs LTS guarantees, native MCP and A2A support as well as EU hosting, the Microsoft Agent Framework 1.0 is the more robust choice. It offers six providers with 1-line swap (Azure OpenAI, OpenAI, Anthropic, Bedrock, Gemini, Ollama) and Azure EU regions (including Frankfurt, West Europe, Switzerland North, Sweden Central).
• Strict sovereignty without a US vendor: Anyone who wants to avoid Microsoft lock-in on principle runs AG2 self-hosted on sovereign DACH infrastructure — or evaluates alternative, likewise self-hostable frameworks.

Rule of thumb: existing v0.2 workloads to AG2. Strategic rebuild within the Microsoft/Azure stack to the Microsoft Agent Framework. Pure AutoGen v0.4 is today mainly the bridge for existing Microsoft AutoGen customers towards the Agent Framework.

Migration: Realistic Paths

The migration path recommended by Microsoft is AutoGen to the Microsoft Agent Framework (recommended for 2026, since AutoGen is run in maintenance mode according to the roadmap). Since AutoGen concepts are absorbed into the Agent Framework's graph workflow, the migration concerns the agents' business logic less than the remodelling of the orchestration as well as tool and provider wrapping.

Anyone who, on the other hand, wants to stay on the classic, conversation-based line has a path with AG2 that does not require a forced migration: the fork explicitly aims at continuity for existing v0.2 codebases. This is the main reason the fork exists at all.

For Agencies and B2B Decision-Makers

For marketing agencies and DACH B2B teams, the lesson from the AutoGen/AG2 split is not "choose repository X", but: choose based on stability, talent pool and hosting, not on GitHub stars. Anyone who has to operate client projects over the long term benefits from the LTS commitment and the EU hosting option of the Microsoft Agent Framework — while consciously accepting the US vendor lock-in and the associated CLOUD Act debate. Anyone who needs maximum independence or an existing v0.2 system without disruption is well served by the community-driven AG2 self-hosted.

As an agency in Vienna, Blck Alpaca supports framework evaluation, GDPR-compliant architecture and the migration of existing AutoGen systems — from the decision matrix to the production multi-agent pipeline. Talk to us before you commit technically: setting the right course at the start saves the expensive re-platforming effort later on.