benchmarks.wiki / Public workspace

GPQA

Science

GPQA is a set of graduate-level questions in biology, physics and chemistry written so that non-experts with web access still fail them. The Diamond subset is the most-used slice.

gpqa

Official source

History

David Rein and collaborators introduced GPQA in 2023 to test expert scientific knowledge that could not readily be substituted by web search. The benchmark focuses on graduate-level biology, physics and chemistry. Its motivation was to construct questions that domain experts could answer but skilled non-experts would struggle with even when allowed to search online.

How it was built

Domain experts wrote the multiple-choice questions, and the collection process included expert validation and attempts by non-experts with web access. This design checks both specialist answerability and resistance to straightforward search. The Diamond subset selects questions using the validation results and is a distinct evaluation slice rather than another name for the whole dataset.

At a glance

What a problem looks like

Graduate-level multiple-choice questions in biology, physics and chemistry.

How it is scored

Access is gated and the authors ask that the questions not be propagated in plain text, to keep them out of training data. This catalogue links to the official distribution instead.

Why it is hard

Written and validated so that skilled non-experts with unrestricted web access still score near chance.

Limitations and criticism

GPQA measures selection among supplied answers in three scientific fields, rather than experimental ability or the production of independently verified research. Results on Diamond should not be conflated with results on other subsets. The authors restrict access and ask that questions not be propagated as plain text to reduce training contamination; those precautions do not establish that any particular model has never encountered the material.

Discussion

1 thread and 0 posts on this page. Anyone may reply; no account is required.

Contribute

Share a reproduction, report an evaluation issue, or suggest a correction to this benchmark overview.

Public problem summary

Catalogue metadata only; public problem statistics are not available.

Public problems
—
Formats
—
Problems with images
—
Answer availability
—
Configs
—
Splits
—
Awaiting a first discussion
—
Problems with a discussion
—
Public contributions
—