skills.rest
sheawinkler
sheawinklerCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

slime-rl-training

Orchestrates RL post-training for GLM models with Megatron-LM and SGLang.

78|16|Updated Apr 23, 2026
Search Tags:#llm#reinforcement-learning#slime#rl-training#sglang#post-training#megatron-lm
One-click install
npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill slime-rl-training-sheawinkler
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: slime-rl-training
Source: https://github.com/sheawinkler/hermes-agent-ultra/tree/main/optional-skills/mlops/slime
Command: npx skills add https://github.com/sheawinkler/hermes-agent-ultra --skill slime-rl-training-sheawinkler

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

slime-rl-training 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.

Core Features & Use Cases

  • RL-oriented post-training orchestration for LLMs
  • Megatron-LM + SGLang integrated training and rollout
  • Data buffering, reward modeling, and evaluation workflows for production RL

Quick Start

Run the slime-rl-training workflow using your Megatron-LM and SGLang setup to begin RL post-training for GLM models.

Frequently Asked Questions about slime-rl-training

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

FAQPage Schema
How do I orchestrate RL post-training for large language models?▼

To orchestrate RL post-training for LLMs, use the slime framework which provides structured guidance for data generation, rollout routing, reward modeling, and policy enforcement. It integrates Megatron-LM and SGLang to handle GLM-scale model training workflows.

Can I use Megatron-LM and SGLang together for LLM rollout and training?▼

Yes, Megatron-LM and SGLang can be used together for integrated training and rollout. The slime framework specifically combines them to enable data buffering, custom generate functions, and evaluation routines within a production RL training stack.

What is needed to configure Megatron-LM for reinforcement learning workflows?▼

Configuring Megatron-LM for RL workflows requires setting up data generation, rollout routing, and reward modeling pipelines. The slime framework satisfies these requirements by providing structured tooling for custom generate functions and evaluation routines across Linux and macOS environments.

Does RL post-training with slime support macOS environments?▼

Yes, RL post-training with slime supports macOS environments. The framework applies data generation, rollout routing, reward modeling, and policy enforcement across both Linux and macOS systems for GLM-scale models.

How do I implement custom data generation workflows for RL training?▼

To implement custom data generation workflows for RL training, use the slime framework's structured tooling. It provides data buffering, custom generate functions, and evaluation routines that integrate with Megatron-LM and SGLang pipelines for production-grade post-training.

Related Skills

Explore More Software Engineering Skills→
jason660519jason660519

slime-rl-training

Integrate Megatron-LM training with SGLang rollout for RL post-training.

Community
Advanced
nelohenriqnelohenriq
1

slime-rl-training

Orchestrates post-training reinforcement learning workflows for LLMs using slime, Megatron-LM integration, and SGLang rollout.

Community
Advanced
ovachieverovachiever
52

training-llms-megatron

Train large language models with Megatron-Core using configurable 3D and 4D parallelism.

Community
Advanced
ZardLi1115ZardLi1115
4

slime-rl-training

Integrate Megatron-LM training with SGLang rollout generation for GRPO-style RL post-training.

Community
Advanced