Saved in:
Bibliographic Details
Main Authors: Gupta, Akanksha, Thomas, Bijo, Asnani, Harshita, Madduru, Phanindra Reddy, Feroze, Samia, Subramanian, Shreyas, Elango, Vikram, Gungor, Mecit
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.05465
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910219653611520
author Gupta, Akanksha
Thomas, Bijo
Asnani, Harshita
Madduru, Phanindra Reddy
Feroze, Samia
Subramanian, Shreyas
Elango, Vikram
Gungor, Mecit
author_facet Gupta, Akanksha
Thomas, Bijo
Asnani, Harshita
Madduru, Phanindra Reddy
Feroze, Samia
Subramanian, Shreyas
Elango, Vikram
Gungor, Mecit
contents As foundation AI models continue to increase in size, an important question arises - is massive scale the only path forward? This survey of about 160 papers presents a family of Small Language Models (SLMs) in the 1 to 8 billion parameter range that demonstrate smaller models can perform as well, or even outperform large models. We explore task agnostic, general purpose SLMs, task-specific SLMs and techniques to create SLMs that can guide the community to build models while balancing performance, efficiency, scalability and cost. Furthermore we define and characterize SLMs' effective sizes, representing increased capability with respect to LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2501_05465
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Small Language Models (SLMs) Can Still Pack a Punch: A survey (updated 2026)
Gupta, Akanksha
Thomas, Bijo
Asnani, Harshita
Madduru, Phanindra Reddy
Feroze, Samia
Subramanian, Shreyas
Elango, Vikram
Gungor, Mecit
Computation and Language
As foundation AI models continue to increase in size, an important question arises - is massive scale the only path forward? This survey of about 160 papers presents a family of Small Language Models (SLMs) in the 1 to 8 billion parameter range that demonstrate smaller models can perform as well, or even outperform large models. We explore task agnostic, general purpose SLMs, task-specific SLMs and techniques to create SLMs that can guide the community to build models while balancing performance, efficiency, scalability and cost. Furthermore we define and characterize SLMs' effective sizes, representing increased capability with respect to LLMs.
title Small Language Models (SLMs) Can Still Pack a Punch: A survey (updated 2026)
topic Computation and Language
url https://arxiv.org/abs/2501.05465