Skip to content
Technology

LoRA (Low-Rank Adaptation)

A lightweight fine-tuning method that adapts large AI models by training only a small number of additional parameters instead of the entire model. Makes custom model training affordable and fast — even on consumer hardware.

Why it matters

LoRA makes customizing an AI model affordable. Instead of retraining billions of parameters, which costs enormous compute, it trains a small add-on, so hobbyists and small teams can teach a model a specific style or task on modest hardware. This is why communities can share thousands of tiny specialized files, letting one base model take on countless personalities and skills without duplicating the whole thing.

A concrete example

An indie illustrator wants an image model to draw in their signature style. They train a LoRA on a few dozen of their own drawings, producing a small file they can layer onto a standard base model. Now the model outputs art in their style on demand. The same idea applies to text models, where a LoRA can specialize one in legal or medical phrasing without a full retrain.

How to use it

This is what makes customising a model practical at small scale: rather than retraining everything, you train a small set of additional parameters, which is dramatically cheaper and produces an adapter you can swap in and out. If you have several related behaviours to support, separate adapters on one base model is usually simpler than several fully fine-tuned models. The method comes from LoRA: Low-Rank Adaptation of Large Language Models.

The common mistake

Expecting an adapter to add knowledge. Like full fine-tuning, it shapes behaviour and style; if the model needs to know your facts, retrieval remains the answer.

Related terms

Put LoRA (Low-Rank Adaptation) into practice

Access 750+ AI models and 60+ tools through Vincony — start free with 100 credits.