verl-rl-training - verl: LLM RL Training FrameworkReinforcement LearningRLHFGRPOPPO★ 1 · Updated 2026-06-30Flexible RL training library for large language models⚙ train models⚙ run rollout⚙ compute reward
