SuperGPQA
A large-scale knowledge-and-reasoning benchmark of ~26,000 graduate-level multiple-choice questions (up to 10 answer options each) spanning 285 academic disciplines, measuring overall answer accuracy.
What this benchmark measures
A large-scale knowledge-and-reasoning benchmark of ~26,000 graduate-level multiple-choice questions (up to 10 answer options each) spanning 285 academic disciplines, measuring overall answer accuracy.
Rows on this page are sourced from public benchmark artifacts, leaderboard exports, or source-linked model reports. Each row keeps benchmark version, source model name, and available run details attached to the score.
The metric shown here is accuracy. It should be interpreted within SuperGPQA, not compared as part of a site-wide ranking.
Frequently asked
What is SuperGPQA?
A large-scale knowledge-and-reasoning benchmark of ~26,000 graduate-level multiple-choice questions (up to 10 answer options each) spanning 285 academic disciplines, measuring overall answer accuracy. It is a reasoning benchmark measured by accuracy.
What does accuracy mean on SuperGPQA?
SuperGPQA reports accuracy (%); higher is better. Scores are shown only within SuperGPQA and are never averaged with other benchmarks.
What is the top reported SuperGPQA score?
Qwen 3.6 Max Preview has the top reported score on SuperGPQA: 73.9% (accuracy).
Why do SuperGPQA scores differ across runs?
Harness, scaffold, reasoning effort, and prompt setup change results, so two runs of the same model can differ. evals.report keeps each score with its run context so the differences stay visible.
Does evals.report rank models across benchmarks?
No. SuperGPQA scores are shown within their own metric; evals.report never combines benchmarks into a composite ranking or a single "best model".