Multi-GPT is an experimental open-source project designed to create GPT-4 models that are fully autonomous. This multi-agent system uses multiple "expertGPTs" that collaborate to perform a task, with each agent having its own short and long-term memory and the ability to communicate with each other. The project employs internet access for searches and information gathering, file storage, and summarization with GPT-3.5, and GPT-4 instances for text generation. Multi-GPT requires Python 3.10 or later, an OpenAI API key (with billing set up), and optional memory backends such as Pinecone, Milvus, or Weaviate. This repository contains extensive documentation and usage guides, along with case studies and resources.
Is there an entry, which should be added to this list?