TRL

HuggingFace Transformer Reinforcement Learning. The gold-standard open library for SFT, PPO, DPO, KTO, and reward modeling. Powers most aligned open-weight models.
Tags
Open Source · Fine-Tuning · RLHF · DPO · PPO · HuggingFace · 18+ · TRL · Free · free tools · free ai tools · no cost