Tag: rl

All the articles with the tag "rl".

• 一个完整的 vime + vLLM RL Training Step

  31 Oct, 2026

  colocate 模式下一次 GRPO 训练步的完整时序图、可调用接口表与参数表，以及按 55 个步骤代入数值、代码和源码位置的逐步讲解。

• Open-Source RL Frameworks for LLMs: verl, OpenRLHF, TRL, NeMo RL, ROLL, slime, AReaL

  29 Oct, 2026

  A side-by-side comparison of the popular LLM RL training frameworks — verl, OpenRLHF, TRL, NeMo RL, ROLL, slime, AReaL — and a deep dive into verl's system design.