OpenManus - An open-source framework for building general AI agents.

## Definition of OpenManus OpenManus is an open-source AI agent framework developed to enable developers to create intelligent task-processing systems efficiently. It emphasizes reinforcement learning-based tuning (e.g., GRPO) to improve large language models (LLMs) in handling complex tasks. The project is community-driven and welcomes contributions. ## Developers of OpenManus OpenManus was developed by contributors from the MetaGPT community, including core authors Xinbin Liang, Jinyu Xiang, Zhaoyang Yu, Jiayi Zhang, and Sirui Hong. The prototype was completed within 3 hours in March 2025. ## Key Features of OpenManus OpenManus offers the following features: - **No invite codes**: Fully open access for all developers. - **LLM API configuration**: Supports models like GPT-4o with customizable parameters (e.g., `max_tokens`, `temperature`). - **Browser automation**: Optional integration with Playwright for enhanced interaction. - **Multi-agent modes**: Includes stable (e.g., `main.py`) and unstable (e.g., `run_flow.py`) modes. - **Community-driven**: Encourages contributions via GitHub issues and pull requests. ## Installation Methods for OpenManus OpenManus requires Python 3.12 and supports two installation methods: 1. **Conda**: ### Citation sources: - [OpenManus](https://github.com/mannaandpoem/OpenManus) - Official URL Updated: 2025-04-01