Back to glossary

Small language model (SLM)

Küçük dil modeli (SLM)

D2 · Generative AI, capabilities and limitations

A small language model is a language model described as relatively compact in parameter count and resource needs for its intended use.

Review status: 2026-12-05

Technical explanation

Small-model reports describe different sizes; this glossary therefore uses no fixed parameter cutoff.

Conceptual boundaries

Model size and execution location are different properties: a small model need not run on a user’s device.

Provider-neutral example

Illustrative: a team evaluates a compact model for classifying short requests within its memory budget.

Limitations

Small does not establish fitness for a task; compare quality and resource use in the intended setting.

Related concepts

Atomic claims and evidence

  1. 1.1Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
    Source
    Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
    Source role
    Authoritative source
    Exact locator
    Abstract: 3.8B phi-3-mini, phone deployment, and 7B phi-3-small
    Supported claim
    Phi-3 reports a 3.8B mini model and a 7B model named small; SmolLM2 describes a 1.7B small language model.
    Last verification
    Review due
    Scope limitation
    These examples illustrate differing size conventions, not a formal size standard.
    1.2SmolLM2: When Smol Goes Big — Data-Centric Training of a Small Language Model
    Source
    SmolLM2: When Smol Goes Big — Data-Centric Training of a Small Language Model
    Source role
    Authoritative source
    Exact locator
    Abstract: 1.7B small language model and data-centric training
    Supported claim
    Phi-3 reports a 3.8B mini model and a 7B model named small; SmolLM2 describes a 1.7B small language model.
    Last verification
    Review due
    Scope limitation
    These examples illustrate differing size conventions, not a formal size standard.