Agent-Tunix Documentation
==========================

Agent-Tunix is a GRPO (Group Relative Policy Optimization) training framework for Gemma3 language models using Google Tunix, with parameter-efficient fine-tuning via LoRA.

.. toctree::
   :maxdepth: 2
   :caption: Getting Started

   getting_started/installation
   getting_started/quick_start
   getting_started/configuration

.. toctree::
   :maxdepth: 2
   :caption: User Guide

   guide/training
   guide/evaluation
   guide/hyperparameter_tuning
   guide/experiments

.. toctree::
   :maxdepth: 2
   :caption: Configuration

   config/overview
   config/model
   config/optimizer
   config/training

.. toctree::
   :maxdepth: 2
   :caption: API Reference

   api/train
   api/evaluate
   api/models
   api/data
   api/rewards

.. toctree::
   :maxdepth: 2
   :caption: Advanced Topics

   advanced/distributed_training
   advanced/custom_rewards
   advanced/troubleshooting

.. toctree::
   :maxdepth: 1
   :caption: References

   references/faq
   references/glossary

Indices and tables
==================

* :ref:`genindex`
* :ref:`modindex`
* :ref:`search`

Quick Links
===========

- :doc:`Quick Start Guide </getting_started/quick_start>`
- :doc:`Configuration Guide </guide/training>`
- :doc:`API Reference </api/train>`
- `GitHub Repository <https://github.com/yourusername/agent-tunix>`_