Multi-GPT

Multi-GPT is an experimental open-source AI system featuring multiple specialized GPT agents that collaboratively work to complete complex tasks. Each agent has its own memory and communication abilities, enabling the system to solve problems dynamically through teamwork. It supports advanced memory management and internet access for information gathering, making it a sophisticated multi-agent AI designed for enhanced problem-solving.

Key Features:

• Collaboration among multiple specialized GPT agents with unique expertise.
• Advanced short-term and long-term memory management for better context retention.
• Internet search and information gathering capabilities for up-to-date data.
• Support for GPT-4 and GPT-3.5 models with flexible backend memory options (e.g., Redis, Pinecone, Milvus, Weaviate).

Use Cases:

• Managing complex research and analytical projects involving diverse information sources.
• Solving multifaceted problems that require coordination of multiple AI components.
• Automating workflows and task management across various domains.

Technical Specifications:

• Uses GPT-4 and GPT-3.5 language models.
• Supports multiple memory backends: local JSON cache, Redis, Pinecone, Milvus, Weaviate.
• Written in Python 3.10+ with integrations for APIs like OpenAI, Google, and ElevenLabs.