OpenRLHF
0.10.2

Getting Started

  • Quick Start

Core Concepts

  • Architecture Foundation: Ray + vLLM Distribution
  • Design Paradigm: Agent-Based Execution

Training Guides

  • RL Training Guide
  • Supervised & Preference Training (SFT / RM / DPO)
  • Common CLI Options

Scaling & Operations

  • Hybrid Engine
  • Async Training & Partial Rollout
  • Performance Tuning
  • Checkpointing
  • Sequence Parallelism (RingAttention)
  • Multi-node Training
  • NVIDIA Docker (Optional)
  • Troubleshooting
OpenRLHF
  • Search

© Copyright 2026, OpenRLHF.

Built with Sphinx using a theme provided by Read the Docs.