slime: An SGLang-Native Post-Training Framework for RL Scaling | LMSYS Org

<svg aria-hidden="true" class="octicon octicon-link" ...

slime: An SGLang-Native Post-Training Framework for RL Scaling - LMSYS Org Projects Blog About Donations Contact ‹ Back to Blog ‹ Back to Blog Contents Vision That Drives slime Customizability Brings Freedom Built for Performance Lightweight and Extensible Roadmap slime: An SGLang-Native Post-Training Framework for RL Scaling The slime Team July 9, 2025 Vision That Drives slime We believe in RL. We believe RL is the final piece toward AGI. If you feel the same way, you'll share our vision: Every field should be end-to-end RLed and every task should become an agent environment. Every RL run sho

related reading

• The upcoming GPT-3 moment for RL | Mechanize, Inc.mechanize.work
• RLHF & Post-Training Course by Nathan Lambertrlhfbook.com
• Composer2.pdfcursor.com
• Keep the Tokens Flowing: Lessons from 16 Open-Source RL Librarieshuggingface.co
• RL at 1T Scale: prime-rl Performance Deep Diveprimeintellect.ai
• How can LLM RL Work Despite Information-Theoretic Inefficiencyberen.io
• State of RL for reasoning LLMs | A. Weersaweers.de
• The Extreme Inefficiency of RL for Frontier Models - Toby Ordtobyord.com
• Reinforcement learning is an infrastructure problemmodal.com
• RL Scaling Laws for LLMs - by Cameron R. Wolfe, Ph.D.cameronrwolfe.substack.com
• Forge: Scalable Agent RL Framework and Algorithm - MiniMax News | MiniMaxminimax.io
• Is Frontier Asynchronous RL Solved? — Luke J. Huangluk-huang.github.io