skills.rest
zhouboyu-xreal
zhouboyu-xrealCOMMUNITY
Explore all skills by this author →
GitHub

slime-rl-training

Scale reinforcement learning post-training with Megatron-LM and SGLang.

Updated Jun 9, 2026
Search Tags:#machine learning#reinforcement learning#model optimization#Megatron-LM#slime framework
One-click install
npx skills add https://github.com/zhouboyu-xreal/Hermes-Memory --skill slime-rl-training-zhouboyu-xreal
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: slime-rl-training
Source: https://github.com/zhouboyu-xreal/Hermes-Memory/tree/main/optional-skills/mlops/slime
Command: npx skills add https://github.com/zhouboyu-xreal/Hermes-Memory --skill slime-rl-training-zhouboyu-xreal

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires sglang-router>=0.2.3, ray, torch>=2.0.0, transformers>=4.40.0, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses the challenge of training AI models with reinforcement learning (RL) post-training, providing a framework for scaling and optimizing machine learning models.

Core Features & Use Cases

  • RL Scaling: Offers tools for scaling reinforcement learning models with the slime framework.
  • Custom Data Generation: Allows for creating custom data generation workflows.
  • Megatron-LM Integration: Provides tight integration with Megatron-LM for training.
  • Use Case: Ideal for research and production environments needing to train GLM models and implement RL for improved performance and scalability.

Quick Start

Run the following command to set up the slime framework for training:

docker pull slimerl/slime:latest
docker run --rm --gpus all --ipc=host --shm-size=16g \
  -it slimerl/slime:latest /bin/bash
cd /root/slime && pip install -e . --no-deps

Frequently Asked Questions about slime-rl-training

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I scale reinforcement learning post-training for large language models?▼

You can scale reinforcement learning post-training using the slime framework, which integrates with Megatron-LM for training and SGLang for high-throughput rollout generation to optimize model performance.

What is the best way to set up a slime framework environment for RL training?▼

To set up the slime framework, pull the slimerl/slime:latest Docker image, run it with GPU support and shared memory, then install the package locally using pip install -e . --no-deps.

Does the slime framework support custom data generation workflows for machine learning?▼

Yes, the slime framework allows for creating custom data generation workflows, providing flexibility in generating training data for GLM models alongside reinforcement learning implementation.

Can I use Megatron-LM and SGLang together for reinforcement learning model optimization?▼

Yes, the slime framework integrates Megatron-LM for training and SGLang for high-throughput rollout generation, enabling combined reinforcement learning post-training and model optimization.

What dependencies do I need for reinforcement learning post-training with the slime framework?▼

Required dependencies include sglang-router, ray, torch, and transformers to execute reinforcement learning post-training and model optimization tasks within the slime framework.

Why use reinforcement learning post-training for GLM models instead of standard training?▼

Reinforcement learning post-training improves performance and scalability for GLM models by leveraging the slime framework's integration with Megatron-LM and SGLang for optimized training and rollout generation.

Related Skills

Explore More Software Engineering Skills→
ldzhhxxldzhhxx
1

slime-rl-training

Run reinforcement learning post-training for LLMs with slime, Megatron-LM, and SGLang.

Community
Advanced
GarrettRoiGarrettRoi

slime-rl-training

Train LLMs with reinforcement learning using Megatron-LM and SGLang.

Community
Advanced
superfhpsuperfhp

slime-rl-training

Automate RL training with Megatron-LM and SGLang rollout generation.

Community
Advanced
LynxLabVNLynxLabVN

slime-rl-training

Automate post-training reinforcement learning optimization for large language models using the slime framework.

Official
Advanced