LIVE
Mistral Makes Large 4 Available as Open Weights29/10/26 · Mistral AI|Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v529/10/26 · Google|Mistral Large 4 now available on Hugging Face29/10/26 · Mistral AI|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|New benchmark checks whether LLMs actually test their temporal medical claims29/10/26|An ODE-based sampling method speeds up discrete diffusion models29/10/26|OpenAI revises its Model Spec with five guiding values for AI behavior29/10/26 · OpenAI|Inside Proton's internal clash over its AI pivot29/10/26 · Proton|Mistral Makes Large 4 Available as Open Weights29/10/26 · Mistral AI|Gemini 4 Carbon tops the Artificial Analysis Intelligence Index v529/10/26 · Google|Mistral Large 4 now available on Hugging Face29/10/26 · Mistral AI|Google Introduces "Gemini 4 Carbon" as Next Step in Its Model Lineup28/10/26 · Google|DeepMind unveils Genie 4, a new generation of world model27/10/26 · Google DeepMind|Anthropic files IPO prospectus, with risk disclosures taking center stage26/10/26 · Anthropic|Robot policy trained only in simulation matches top real-world systems26/10/26|OpenAI unveils GPT-6.1 Terra, a model built for computer use26/10/26 · OpenAI|New benchmark checks whether LLMs actually test their temporal medical claims29/10/26|An ODE-based sampling method speeds up discrete diffusion models29/10/26|OpenAI revises its Model Spec with five guiding values for AI behavior29/10/26 · OpenAI|Inside Proton's internal clash over its AI pivot29/10/26 · Proton|
AppRisingofficial publisher

AutoGen

App : A product you use directly

Microsoft's multi-agent framework, a pioneer in the field.

In maintenance mode: Microsoft now pushes Microsoft Agent Framework, do not start a new project on it.

GitHub · 61,550 ★
Overall score
7.9 /10
Security: nothing to flag
Utility
6.0
Reliability
10.0
Security
No caveat on record
Ease
Technical · Technical

Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.

What it does, concretely

Coordinating multiple AI agents to accomplish complex tasks requires solid infrastructure: message handling between agents, workflow orchestration, and connections to various language models. AutoGen addresses this need with a Python (and .NET) framework designed to build multi-agent applications that can act autonomously or alongside humans.

The framework is built on a layered architecture: a Core API for message passing and event-driven agents, an AgentChat API for rapidly prototyping common scenarios such as two-agent or group chats, and an Extensions API for connecting LLM clients (OpenAI, Azure OpenAI) and additional capabilities like code execution. A no-code tool, AutoGen Studio, allows building multi-agent workflows through a graphical interface.

AutoGen is aimed at Python or .NET developers who want to experiment with multi-agent architectures, particularly those who already have projects based on version 0.2 of the framework and need to maintain them. It requires Python 3.10 or later.

For any new project, AutoGen is no longer the right starting point: Microsoft now maintains it in community mode without adding new features, and officially recommends migrating to Microsoft Agent Framework, its long-term supported successor.

What it offers

  • ·Multi-agent orchestration via the AgentChat API, supporting common patterns such as two-agent or group chats.
  • ·Core API for message passing between agents and event-driven agent management, locally or in a distributed setup.
  • ·Extensions API for connecting LLM clients (OpenAI, Azure OpenAI) and adding capabilities such as code execution.
  • ·Compatibility with the MCP (Model Context Protocol) to connect external servers, for example Playwright for web browsing.
  • ·AutoGen Studio: a no-code graphical interface for prototyping and running multi-agent workflows locally.
  • ·AutoGen Bench: a benchmarking suite for evaluating agent performance.
  • ·Python and .NET support, with interoperability between both environments via the Core API.

In practice

Price
Free
Commercial use
Commercial use: yes
Available on
cli
Licence
MIT

Security and data

Rating
Security: nothing to flag
Your data
Data stays local
Known incidents
Projet passé en maintenance en 2026.
Maintenance
Active
Last release
30 September 2025

Links and repository checked on 25 October 2026 · added on 1 September 2023

How to install AutoGen

1. Make sure Python 3.10 or later is installed on your machine. 2. Install the AgentChat package and OpenAI client via pip: pip install -U "autogen-agentchat" "autogen-ext[openai]". 3. Export your OpenAI API key in your terminal: export OPENAI_API_KEY="sk-...". 4. (Optional) Install AutoGen Studio for the no-code interface: pip install -U "autogenstudio". 5. (Optional) Launch AutoGen Studio with: autogenstudio ui --port 8080 --appdir ./my-app. 6. Only connect to trusted MCP servers, as they can execute commands in your local environment.
pip install -U "autogen-agentchat" "autogen-ext[openai]"

In the same category

Whole category Orchestrate →