Skip to main content

Verl Rl Training

Skill Verified Active

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.

Purpose

To enable users to effectively train large language models at scale using reinforcement learning techniques with the verl framework, offering production-ready guidance and support.

Features

  • Guidance on verl RL training library
  • Support for RLHF, GRPO, PPO, and other RL algorithms
  • Flexible infrastructure backend configurations (FSDP, Megatron)
  • Detailed troubleshooting and common issue resolution
  • Examples for various training workflows

Use Cases

  • Implementing RLHF for LLM post-training
  • Training LLMs at scale with flexible infrastructure
  • Leveraging GRPO for math and reasoning tasks
  • Configuring PPO with a critic model for dense reward tasks

Non-Goals

  • Implementing Megatron-native training directly (use slime or miles)
  • Simple SFT/DPO tasks (use TRL or Axolotl)
  • Core LLM architecture development
  • Basic language model fine-tuning

Installation

First, add the marketplace

/plugin marketplace add Orchestra-Research/AI-Research-SKILLs
/plugin install AI-Research-SKILLs@ai-research-skills

Quality Score

Verified
99 /100
Analyzed about 22 hours ago

Trust Signals

Last commit16 days ago
Stars8.3k
LicenseMIT
Status
View Source

Similar Extensions

Verl Rl Training

95

Provides guidance for training LLMs with reinforcement learning using verl (Volcano Engine RL). Use when implementing RLHF, GRPO, PPO, or other RL algorithms for LLM post-training at scale with flexible infrastructure backends.

Skill
davila7

Openrlhf Training

99

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

Skill
Orchestra-Research

Slime Rl Training

98

Provides guidance for LLM post-training with RL using slime, a Megatron+SGLang framework. Use when training GLM models, implementing custom data generation workflows, or needing tight Megatron-LM integration for RL scaling.

Skill
Orchestra-Research

Openrlhf Training

97

High-performance RLHF framework with Ray+vLLM acceleration. Use for PPO, GRPO, RLOO, DPO training of large models (7B-70B+). Built on Ray, vLLM, ZeRO-3. 2× faster than DeepSpeedChat with distributed architecture and GPU resource sharing.

Skill
davila7

Torchforge

96

Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.

Skill
Orchestra-Research

Fine Tuning With Trl

96

Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.

Skill
Orchestra-Research

© 2025 SkillRepo · Find the right skill, skip the noise.