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.
- Score
- Unranked
- Evidence
- No signals yet
- Last commit
- Not known
- Listed
Install
armory install slime-rl-training --cli claudewrites the skill into.claude/skills/slime-rl-training/SKILL.mdListed as compatible
# fetches the source and writes it to:
.claude/skills/slime-rl-training/SKILL.mdNeeds the armory CLI · not on npm yet, build it from cli/ in the repository
What it is
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.
When to use it
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.
How to install / invoke
# Copy the skill into your .claude/skills/ directory
curl -sL https://raw.githubusercontent.com/davila7/claude-code-templates/main/cli-tool/components/skills/ai-research/post-training-slime/SKILL.md -o .claude/skills/post-training-slime/SKILL.md
Notes
Extracted from davila7/claude-code-templates, post-training-slime category.