A field guide to AI positions

AI scaling laws

Sources are over 18 months old.

Publication dates and source age

Sources counted: 2

Newest dated source: 2022-03-29

Assessed at this edition's evidence cutoff: 2026-09-15. 18-month boundary: 2025-03-15.

Publication age does not tell us whether a claim is still valid. Reading an old source again does not make its publication date newer. An update date does not establish that the passage we used was updated.

Open in the glossary Reading notes · Structured record

In plain language

Measured patterns linking model size, training data and computing resources to performance. Each pattern concerns a particular measurement and training setup. [1] [2]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

Limits & distinctions

Larger is not the only choice. Hoffmann and coauthors found better results by balancing model size with more training data. Extrapolating a measured trend to untested scales adds an assumption. [1] [2]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

A fuller explanation

Researchers train different-sized models and fit equations to the results. In language-model research, a common measure is prediction loss: how poorly the model predicts the next text piece. [1] [2]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

How it relates to the map

Map context: these studies inform expectations about capability growth and resource use. [1] [2]

Reference this explanation or suggest a correction

Link to this explanation · Suggest a correction · How corrections work

Share this page

https://theaiatlas.org/ideas/scaling-laws/

Download a share image · Vector image

Image previews are summaries. Keep the page link so readers can check the evidence.

Sources and what we read

  1. 1. Scaling Laws for Neural Language Models

    Source is over 18 months old.

    Publication dates and source age

    Sources counted: 1

    Newest dated source: 2020-01-23

    Assessed at this edition's evidence cutoff: 2026-09-15. 18-month boundary: 2025-03-15.

    Publication age does not tell us whether a claim is still valid. Reading an old source again does not make its publication date newer. An update date does not establish that the passage we used was updated.

    Read abstract and submission record. Empirical relationships concern language-model prediction loss and training resources, not an AGI date.

  2. 2. Training Compute-Optimal Large Language Models

    Source is over 18 months old.

    Publication dates and source age

    Sources counted: 1

    Newest dated source: 2022-03-29

    Assessed at this edition's evidence cutoff: 2026-09-15. 18-month boundary: 2025-03-15.

    Publication age does not tell us whether a claim is still valid. Reading an old source again does not make its publication date newer. An update date does not establish that the passage we used was updated.

    Read abstract and submission record. Tests balance parameter count and training tokens under a fixed compute budget. Results are scoped to the studied setups.

Edition and machine-readable evidence

Content version 0.20.0. Evidence cutoff 2026-09-15; this does not mean every source was read on that day.

Pinned complete dataset · Complete evidence page · Agent consumption guide