Small Language Model (SLM)
Also called: SLM
A small language model is a compact model chosen in place of a frontier one because it costs less to run, answers faster and is small enough to host on hardware you control. The trade is breadth for control. A small model is pointed at a narrow, well-defined job, classifying documents, extracting fields from forms, answering questions about one product line, and can be fine-tuned until it performs well inside that job while remaining weaker at anything outside it.
In practice
Where a task is repetitive and high volume, a small model is frequently the cheaper answer, and running it on your own infrastructure keeps the data inside the perimeter. The cost moves rather than disappears: you take on the hosting, the fine-tuning and the work of keeping the model current.