🧰 Verl

The items that name it 6

• Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
  writeup2026-10-22

  RL training framework, version 0.9.0, used with a patched checkout to run GRPO.

• [AINews] not much happened today
  article2026-10-20

  RL framework moving toward async agent RL; delta weight synchronization cut weight-transfer time 10.3x.

• [AINews] not much happened today
  article2026-10-20

  RL training framework moving toward async agent RL; delta weight synchronization cut weight-transfer time 10.3x.

• [AINews] Agents disprove Polysemanticity; greatly improve Interpretability & Alignment
  article2026-10-16

  Training framework for LoRA GRPO

• ClusterMAX 3.0: The Industry Standard GPU Cloud Rating System Returns
  article2026-09-23

  Open-source RL post-training framework already run in PostTrainingX.

• baidu-baige/LoongSage
  repo2026-08-28

  Provided an excellent design paradigm for an extensible RL training and inference framework, which inspired LoongSage