AutoGen: Enable Next-Gen Large Language Model Applications

AutoGen

Enable Next-Gen Large Language Model Applications

AutoGen is a platform that supports the creation of LLM applications with multiple agents that can communicate with each other to accomplish tasks. AutoGen agents are adaptable, interactive, and smoothly integrate human involvement. They can work in different modes that use combinations of LLMs, human inputs, and tools.

Visit Website

Features

Getting Started 🚀

If you want to get started right away, follow these simple steps:

1. Click on the button below to launch the GitHub Codespace

2. Rename OAI_CONFIG_LIST_sample to OAI_CONFIG_LIST and move it to /notebook folder. Make sure to adjust the config settings as needed.

3. Enjoy exploring the notebooks!

AutoGen is a framework for creating LLM applications with conversational agents and minimal coding

AutoGen supports various conversation patterns and agent configurations for complex workflows

AutoGen showcases a range of working systems from different domains and complexities

AutoGen offers an enhanced inference API with performance tuning, utilities, and advanced features

AutoGen is based on collaborative research from Microsoft, Penn State University, and the University of Washington

Installation

To use AutoGen, you need Python version 3.8 or higher. You can get it from pip:

pip install pyautogen

This will install the basic dependencies. For additional features, you can specify extra options when installing.

See Installation for more options.

We highly recommend using docker and the python docker package for code execution.

For LLM inference settings, refer to the FAQs.

from autogen import AssistantAgent, UserProxyAgent, config_list_from_json
# Load LLM inference endpoints from an env variable or a file
# See https://microsoft.github.io/autogen/docs/FAQ#set-your-api-endpoints
# and OAI_CONFIG_LIST_sample
config_list = config_list_from_json(env_or_file="OAI_CONFIG_LIST")
# You can also set config_list directly as a list, for example, config_list = [{'model': 'gpt-4', 'api_key': '<your OpenAI API key here>'},]  
assistant = AssistantAgent("assistant", llm_config={"config_list": config_list})
user_proxy = UserProxyAgent("user_proxy", code_execution_config={"work_dir": "coding"})
user_proxy.initiate_chat(assistant, message="Plot a chart of NVDA and TESLA stock price change YTD.")
# This initiates an automated chat between the two agents to solve the task

Multi-Agent Conversation Framework

Autogen is a generic framework for building next-gen LLM applications that involve multi-agent conversations. It allows users to create and customize conversable agents that can integrate LLMs, tools, and humans. By enabling chat among multiple intelligent agents, Autogen can facilitate tasks that require collective problem-solving, automation, or human guidance, including tasks that involve using tools via code.

Some of the features of this use case are:

✅Multi-agent conversations

Autogen enables communication among different agents to accomplish tasks. This enables more advanced and diverse applications than what a single LLM can do.

✅Customization

Autogen allows users to tailor the agents to their specific application needs. This includes the option to select the LLMs, the human input modes, and the tools to use.