slime is a lightweight RL post-training framework that connects Megatron training with SGLang rollout. It has been validated through GLM model releases and addresses the problem of disconnected trainers, rollout services, and agent frameworks.
Project overview
The framework is battle-tested by frontier model training as the RL framework behind the GLM family, offering an opinionated and deeply focused Megatron + SGLang path for large-scale reinforcement learning.
Project type
AI Agent · Model Development · Model Runtime
Use cases
Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
AI engineers and researchers performing large-scale RL post-training who want an opinionated Megatron + SGLang path.
Key capabilities
Supports efficient training in various modes by connecting Megatron with SGLang.
Enables arbitrary training data generation workflows through custom data generation interfaces and server-based engines.
Passes Megatron arguments through directly and exposes installed SGLang arguments with a --sglang- prefix.
Supports PD Disaggregation for multi-turn and agentic workloads with different prefill/decode resource needs.
Ships examples for multi-agent rollout, search/RAG-style multi-turn generation, fully-async rollout, and SWE coding-agent RL.
Limitations and risks
Optimizes deeply for SGLang as the single rollout backend instead of supporting multiple inference engines.
Getting started
Setup is hard, requiring understanding of RL, Megatron, and SGLang configuration via command-line arguments, along with environment setup and data preparation before starting training.
Evidence and sources
GitHub project description: slime is an LLM post-training framework for RL Scaling.
README: slime's design goal is to make these two capabilities reinforce each other without turning the system into a heavy stack of disconnected trainers, rollout services, and agent fram…
README: slime is the RL framework behind GLM-5.2, GLM-5.1, GLM-5, GLM-4.7, GLM-4.6, and GLM-4.5.
README: Lightweight and opinionated: slime focuses deeply on the Megatron + SGLang path used for large-scale RL.