Skip to content

Agent Lightning

Agent Lightning is the absolute trainer to light up AI agents.

Join our Discord community to connect with other users and contributors.

Features

  • Turn your agent into an optimizable beast with ZERO CODE CHANGE (almost)! 💤
  • Build with ANY agent framework (LangChain, OpenAI Agent SDK, AutoGen, CrewAI, Microsoft Agent Framework...); or even WITHOUT agent framework (Python OpenAI). You name it! 🤖
  • Selectively optimize one or more agents in a multi-agent system. 🎯
  • Embraces Algorithms like Reinforcement Learning, Automatic Prompt Optimization, Supervised Fine-tuning and more. 🤗

How to Read this Documentation

This documentation is organized into the following parts:

  • Installation - Get started with Agent Lightning
  • How-to Recipes (e.g., Train SQL Agent with RL) - Practical examples of training agents and customizing algorithms.
  • Learning More (e.g., Debugging) - Guides on specific topics like debugging or parallelization.
  • Algorithm Zoo (e.g., APO) - References for built-in algorithms.
  • Deep Dive (e.g., Bird's Eye View) - For a deeper understanding of what Agent-lightning is doing under the hood.
  • API References (e.g., Agent) - References for the Agent-lightning Python API.

Resources

Community Projects

  • DeepWerewolf — A case study of agent RL training for the Chinese Werewolf game built with AgentScope and Agent Lightning.
  • AgentFlow — A modular multi-agent framework that combines planner, executor, verifier, and generator agents with the Flow-GRPO algorithm to tackle long-horizon, sparse-reward tasks.

Citation

If you find Agent Lightning useful in your research or projects, please cite our paper:

@misc{luo2025agentlightningtrainai,
      title={Agent Lightning: Train ANY AI Agents with Reinforcement Learning},
      author={Xufang Luo and Yuge Zhang and Zhiyuan He and Zilong Wang and Siyun Zhao and Dongsheng Li and Luna K. Qiu and Yuqing Yang},
      year={2025},
      eprint={2508.03680},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2508.03680},
}

License

See the LICENSE file for details.