THUDM/slime
Star THUDM / slime slime is an LLM post-training framework for RL Scaling.
Python8.2K stars1.2K forks
What it does
slime is a powerful LLM post-training framework designed for reinforcement learning scaling, enabling efficient training and flexible data generation. It supports various models and facilitates advanced research in AI.
Star history
Not enough history yet — 1 day(s) recorded. The daily snapshot builds this up.
Tracking
- Last trending
- 2026-02-21
Creator kit
Hook
Unlock the potential of AI with slime, the ultimate framework for high-performance reinforcement learning training!
Content angles
- Create a tutorial series on setting up and using slime for RL projects.
- Write a blog post comparing slime with other RL frameworks and its unique features.
- Develop a video showcasing successful projects built with slime, highlighting their impact.
Who should care
AI researchers, machine learning engineers, and developers interested in reinforcement learning.