gpqa-diamond/gpqa-diamond
Science
Biology
Chemistry
Physics
Knowledge
GPQA Diamond: expert-validated graduate-level multiple-choice questions in biology, physics, and chemistry, designed to be Google-proof for non-experts.
Run this task
CLI:
inspect eval inspect_harbor/gpqa_diamond --model openai/gpt-5Python:
from inspect_ai import eval
from inspect_harbor import gpqa_diamond
eval(gpqa_diamond(), model="openai/gpt-5")Dataset information
| Harbor registry | gpqa-diamond/gpqa-diamond |
| Inspect task | gpqa_diamond |
| Latest digest | sha256:87d299c2cd19daac11ff3e8f3dfb61f614cad21a8f93f22e188f4b11282dfcf1 |
| Samples | 198 |
| Paper | arxiv |
| Source | https://github.com/idavidrein/gpqa |
See Task Parameters for the parameter set shared across all Harbor tasks.