verifiers: Reinforcement Learning with LLMs in Verifiable Environments

Prime Intellect's open-source Python library for building verifiable RL environments and training LLMs with GRPO-style algorithms.