AutoGen is a framework that allows the development of LLM (large language model) applications using multiple agents that can communicate with each other to solve tasks. These agents are customizable, conversable, and can involve human input. AutoGen simplifies complex LLM workflows and maximizes the performance of LLM models. It supports a variety of conversation patterns and provides working systems across different applications and complexities. AutoGen also offers enhanced inference capabilities for LLMs, such as tuning, caching, error handling, and templating. It is a collaborative project involving Microsoft, Penn State University, and the University of Washington. Documentation and code examples are available for further exploration.
https://github.com/microsoft/autogen