跳转至

BEAM: Token-efficient Multi-agent Inference Framework

BEAM is a high-performance distributed multi-agent reasoning framework that focuses on maximizing token utilization and reasoning efficiency while ensuring reasoning quality.

🚀 Key Features

  • Getting Started

    Install BEAM and train your first topic model in minutes

    Quick Start

  • Models

    Understand the workflow and structure

    User Guide

  • Configuration

    Basic configuration settings

    Models

  • Examples

    Model usage examples

    API Reference

Citation

If you use BEAM in your research, please cite:

@software{beam2024,
  title={BEAM: A Bayesian Energy-Aware Framework for Multi-Agent Communication Optimization under Incomplete Information},
  author={BEAM Team},
  year={2024},
  url={https://github.com/erwinmsmith/BEAM}
}
  • If this is your first time using it, please check out quickstart。
  • For a deeper understanding of the system design, please read core architecture。
  • For function details, please refer to the API Reference。