AI Glossary Term
What does “Small Language Model (SLM)” mean?
Definition
A compact language model optimized for lower latency, cost, or on-device usage.
How Small Language Model (SLM) works
Example
A note-taking app runs a compact model directly on the phone to summarize short memos. It responds quickly, works offline, and keeps the text on the device, though it handles hard requests less well than a large cloud model.
Why it matters
Smaller models cut cost, delay, and privacy exposure. They are often a good fit for narrow tasks.
Common questions
How small is small?
There is no fixed line. The term is commonly used for models from a few hundred million to a few billion parameters.
Are they worse?
Usually less capable in general, but a small model tuned for one task can match larger ones on that task.
How are they made?
By training smaller architectures, or shrinking bigger ones through distillation, pruning, or quantization.