Training
LoRA (low-rank adaptation)
A cheap fine-tuning technique that trains a small set of extra weights instead of updating the whole model.
LoRA adapters are small files that can be swapped in and out of a base model, which makes it practical to keep many specialised versions of one open-weight model.
Frequently asked questions
What does LoRA (low-rank adaptation) mean?
A cheap fine-tuning technique that trains a small set of extra weights instead of updating the whole model. LoRA adapters are small files that can be swapped in and out of a base model, which makes it practical to keep many specialised versions of one open-weight model.