AI Governance Engineer
GitHub

Small language model (SLM)

By Updated

A language model small enough to run close to the user, for example on a phone 1. Its controls must ship with it: guardrails on the device, a version inventory across the fleet and a kill switch delivered as a remote flag or app update.

Developed in
ch. 11, LLMs and SLMs
Chapters
ch. 11, AI Defined
Contrast with
Large language model (LLM)
Source
1 numbered reference, listed below

Where it is used

One chapter of the Body of Knowledge uses the term. Each link opens the first section that does.

  • Kill switch

    A tested mechanism to stop an agent or system from acting; a precondition of granting autonomy, registered against the agent's identity.

  • Guardrail

    A runtime control that inspects or mediates a model's or agent's inputs, outputs or tool calls and blocks, rewrites or escalates what breaks a policy, logging…

Sources

  1. [1] Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone (3.8-billion-parameter model deployable on a phone; arXiv 2404.14219). Abdin et al. (Microsoft). 2024-04-22. https://arxiv.org/abs/2404.14219 (verified: primary)

Definitions of legal terms paraphrase the cited text, which governs. Dated statements are as of .

Cite this term

García Aibar, J. (2026). Small language model (SLM). In AI Governance Engineering: The Thesis & Body of Knowledge (v0.5.0), Glossary. https://doi.org/10.5281/zenodo.22956197. https://aigovernanceengineer.com/glossary/small-language-model-slm. CC BY 4.0

BibTeX

@misc{aige2026smalllanguagemodelslm,
  author  = {Jorge García Aibar},
  title   = {{Small language model (SLM)}},
  note    = {Glossary, AI Governance Engineering: The Thesis \& Body of Knowledge, version 0.5.0},
  year    = {2026},
  doi     = {10.5281/zenodo.22956197},
  url     = {https://aigovernanceengineer.com/glossary/small-language-model-slm}
}