ChatLearn Documentation

Introduction

• ChatLearn: A flexible and efficient reinforcement learning framework for large language models(LLMs)

Installation

• Environment and Code Preparation

Tutorial

• End-to-End GRPO Training Tutorial with FSDP
• End-to-End GRPO Training Tutorial with Mcore
• Multi-Node Distributed Training
• Resume Training and Fault Tolerance
• Performance Tuning Guide
• Profile

Customized Task

• Dataset Preparation
• Customize Reward Function

Configuration

• Config Explanation

FAQ

• FAQ
• Common Errors