MetaGPT (Hong et al.) is an awesome multi-agent framework (ICLR 2024) that models agents as a software company following a structured SOP – PMs, architects, engineers, and more communicate with each other to solve the task at hand.
LLamaIndex team excited to feature RAG-enhanced MetaGPT, powered by @llama_index modules – this allows agents to tap into domain-specific knowledge.
MetaGPT Features
Features:
- Data input, supporting diverse file formats and Python objects.
- Retrieval functionality, compatible with Faiss, BM25, ChromaDB, and Elasticsearch, as well as hybrid retrieval methods.
- Post-retrieval processing, featuring LLM Rerank and ColbertRerank options to reorder retrieved content for enhanced accuracy.
- Data updating capabilities, allowing the addition of text and Python objects.
- Data storage and recovery without requiring vectorization on each access
Script: https://github.com/geekan/MetaGPT/blob/main/examples/rag_search.py
Paper: https://arxiv.org/pdf/2308.00352v5.pdf
Read related articles: