Government · Government · NIST

Practices for Automated Benchmark Evaluations of Language Models

29 Jan 2026

Read Source at nvlpubs.nist.gov ↗Read herePublic domain (US government work)

Outline · 39 pages

This document has no section headings. Open at p. 1

How to cite

A citation names the document, the version and the place.

nist:nist.ai.800-2.ipd/r1/p1

https://safetypage.noahairmet.com/d/nist/nist.ai.800-2.ipd/r1/p1

BibTeX

@misc{nist2026,
  title = {Practices for Automated Benchmark Evaluations of Language Models},
  author = {{NIST}},
  year = {2026},
  url = {https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.800-2.ipd.pdf}
}

Versions

Version	Published	Read
r1 latest	29 Jan 2026	Here

Licence

Terms: Public domain (US government work)
So you: Read the full document here
Evidence: Statute: 17 U.S.C. 105: copyright protection is not available for any work of the United States Government. https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.800-2.ipd.pdf — NIST publication; US federal government work, public domain under 17 U.S.C. §105 ("This publication is available free of charge from https://doi.org/10.6028/…"; NIST works are not subject to US copyright).
Read at: www.law.cornell.edu/uscode/text/17/105, 10 Nov 2026