Hugging Face PEFT (Parameter-Efficient Fine-Tuning) allows fine-tuning of large Transformers Architecture models with fewer parameters, optimizing both cost and efficiency.

https://huggingface.co/docs/transformers/peft