Explore projects

P1 Base ykeohane

A set of examples based on verl for end-to-end RL training recipes.

verl: Volcano Engine Reinforcement Learning for LLMs

Training VLM agents with multi-turn reinforcement learning