GPQA Diamond.

GPQA Diamond is a dataset for language modeling [1]. It consists of 448 expert-validated multiple-choice questions in STEM fields. It is designed to be a challenging benchmark for advanced AI reasoning and drives progress in scalable oversight and structured problem-solving [2].

Best published scores.

No results indexed yet — be the first to submit a score.

Neighbouring benchmarks.

Other Language Modeling datasets on Codesota.

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

• 01A public checkpoint or API endpoint
• 02A reproduction script with frozen commit + seed
• 03Declared evaluation environment (Python, deps)
• 04One row per metric declared by this dataset
• 05A contact so we can follow up on discrepancies