{"data":{"slug":"fine-tuning","etype":"concept","name":"Fine-tuning & LoRA","cat":"Training & Inference","subtitle":"Adapting a trained model cheaply — the small-compute on-ramp.","summary":"Fine-tuning continues training on a narrow dataset to specialize a model; LoRA does it by learning tiny low-rank adapter matrices instead of touching all weights, cutting cost by orders of magnitude. It is how enterprises adapt open models, how the Stable Diffusion style ecosystem works, and what products like Tinker sell as a service.","attrs":null,"kf":null,"aliases":["LoRA","PEFT"],"ticker":null,"exchange":null,"hq":null,"founded":null,"website":null,"status":null,"region":null,"asof":"2026-01-15","sources":[{"url":"https://arxiv.org/abs/2106.09685","label":"LoRA paper"}]},"meta":{"found":true,"count":1,"served_from":"site cache (no live upstream calls)","as_of":"2026-10-14T04:31:02+00:00","disclaimer":"Research and information only — not investment advice. Structured records carry basis labels (vendor spec / reported / estimate); market quotes are delayed/indicative.","attribution":"Records compiled from vendor documentation, official filings and reputable public reporting; market quotes via financial data providers. See /data-sources."}}