TRL
Hugging Face library for training language models with RLHF, SFT, and DPO
TRL (Transformer Reinforcement Learning) is the standard Hugging Face library for fine-tuning language models. It supports supervised fine-tuning (SFT), reinforcement learning from human feedback (RLHF), direct preference optimization (DPO), and other alignment techniques. Built on top of Transformers and integrates with PEFT for parameter-efficient training.
Pricing: Free
TRL Alternatives
Explore 21 products in the Fine-tuning category. View all TRL alternatives.
Replicate
Run and fine-tune open-source models. Deploy custom models at scale. All with one line of code.
fal
Build the next generation of creativity with fal. Lightning fast inference.
Work on TRL? Feature it at the top of Fine-tuning.
Is your product missing?