Skip to content
ai

PEFT

Parameter-Efficient Fine-Tuning

Definition

PEFT refers to a family of fine-tuning techniques that update only a small fraction of a pre-trained model's parameters while achieving performance comparable to full fine-tuning. Methods include LoRA, adapters, prefix tuning, and prompt tuning.

PEFT dramatically reduces compute, memory, and storage requirements, making fine-tuning accessible on consumer hardware.

Related Terms

LoRA
Low-Rank Adaptation
QLoRA
Quantized Low-Rank Adaptation
Adapter
Adapter Layer
Fine-tuning
Model Fine-tuning
← Back to Glossary

Ship secure code faster

Crash Override integrates security into the developer workflow. No context switching, no waiting on reviews.

Talk to a Human See the Product